Unspecified AI system

ChatGPT’s expanding world of work

Explore OpenAI Signals data on 165 work activities across nearly 100 countries, revealing how people use ChatGPT for work and how usage changes over time.

OtherprovisionalTier 2 Significant Harm−0.17

How an AI start-up founded by ex-DOGE staffers broke into Anthropic - The Washington Post

How an AI start-up founded by ex-DOGE staffers broke into Anthropic The Washington Post

OtherprovisionalTier 2 Significant Harm−0.25

Anthropic’s Claude bypasses paywall to access Georgia voter data without authorization - TradingView

Anthropic’s Claude bypasses paywall to access Georgia voter data without authorization TradingView

Daily briefing: The Selfish Gene, 50 years on

Milestone cancer vaccine success leaves hard questions unanswered

mRNA vaccine from Merck and Moderna slashes risk of melanoma’s return, but why remains unclear

Anthropic’s delay in reporting ‘rogue agent’ Georgia vote hack alarms government officials - New York Post

Anthropic’s delay in reporting ‘rogue agent’ Georgia vote hack alarms government officials New York Post

STAT+: Nonprofit Bon Secours Mercy Health strikes $5.7 billion deal for surgery center chain

The deal would make the Catholic system one of the largest operators of outpatient surgery centers.

Mentoring a lab colleague? Share your ‘CV of failures’ with them

Nerve activity initiates autoimmune hair loss

STAT+: Rising premiums, a sad Medicaid gathering, and a new fraud crackdown

The cost of health insurance is surging, and CMS presses ahead with a new fraud crackdown.

SafetyprovisionalTier 2 Significant Harm−0.07

Claude’s Voter-Data Incident Raises Fresh Questions over US AI Security Oversight - تسنیم

Claude’s Voter-Data Incident Raises Fresh Questions over US AI Security Oversight تسنیم

STAT+: Incyte pancreatic cancer medicine induces lasting tumor responses, company says at ESMO

A targeted drug from Incyte, when used in combination with chemotherapy, induced lasting tumor responses in certain patients with pancreatic cancer

OtherprovisionalTier 2 Significant Harm−0.05

Harm21 Oct 2026

AI doctors’ orders come with a sky-high bill

Benefit21 Oct 2026

Teen depression and anxiety predicted by childhood brain development

Benefit21 Oct 2026

Surprise! White dwarf stars come in a rainbow of colours

Benefit21 Oct 2026

Can plants defend coasts against storms?

Benefit20 Oct 2026

This ‘glue gun ant’ fires sticky webs at its enemies

Lordomyrma azumai has repurposed its venom gland as a web slinger

Why failure is a taboo topic in science

Opinion: My mother killed herself. I wish assisted suicide had been an option

“Sometimes mental illness is terminal,” Torie Bosch writes in a personal essay about medical aid in dying for psychiatric reasons.

STAT+: Top immunology association divided over $8.4-billion California ballot measure to fund the field, emails show

Internal emails obtained by STAT show that the American Association of Immunologists is sharply divided over California's Proposition 38, even though it would provide $8.4 billion in taxpayer dollars for…

How pregnancy reshapes the brain throughout life

Medicine & HealthprovisionalTier 2 Significant Harm−0.05

Harm18 Oct 2026

AI Breach of Australian Healthcare Website Sparks Regulation Debate - Carolina Political Review

AI Breach of Australian Healthcare Website Sparks Regulation Debate Carolina Political Review

Harm18 Oct 2026

OpenCode Security Breach: Distillation Attack Hit Console Users With Linked AI Providers. How to Find OpenCode on Developer Machines and Respond - Step Security

OpenCode Security Breach: Distillation Attack Hit Console Users With Linked AI Providers. How to Find OpenCode on Developer Machines and Respond Step Security

SafetyprovisionalTier 2 Significant Harm−0.12

STAT+: Pharmalittle: We’re reading about an HHS peptide push, Trump’s pharma stock holdings, and more

The Trump administration is proceeding with a plan to allow compounding pharmacies to produce certain peptides

OtherprovisionalTier 1 Limited Harm−0.00

Harm16 Oct 2026

PhD students who write dissertation with AI less likely to stay in academia

Harm16 Oct 2026

OpenAI discloses four more rogue incidents involving financial, medical, and government data - Fortune

OpenAI discloses four more rogue incidents involving financial, medical, and government data Fortune

SafetyprovisionalTier 2 Significant Harm−0.25

How Itaú is building generative AI for its entire customer base

With ia.i, a conversational interface, and the IARA governance platform, Itaú is applying its AI First strategy to customers and employees — in one of Brazil’s most regulated sectors.

History’s long-secret codes are being cracked by amateurs using AI

Memory stays strong in old age when this brain enzyme is active

Daily briefing: We’re close to reaching a 30% land-conservation target

Benefit12 Oct 2026

Immorta Bio says AI helped identify SenoMune-0257, a nine-amino-acid senolytic peptide

On 12 October 2026 Immorta Bio announced it had finished planned preclinical studies of SenoMune-0257, a synthetic nine-amino-acid peptide it says AI helped select to stimulate T cells against senescent cells. The company aims for first human evaluation before year-end 2026. No sequence, named model, public data package, or trial site was disclosed.

Benefit7 Oct 2026

Retrofitted LLM can count the letter ‘i’s in ‘artificial intelligence’

Benefit7 Oct 2026

‘When prevention disappears, infections rise’: can the world still end AIDS?

Benefit6 Oct 2026

Scientists take a radiometer for a spin in a hot-air balloon

Harm2 Oct 2026

OpenAI is facing a lawsuit after a hack carried out by an AI agent - صوت الإمارات

OpenAI is facing a lawsuit after a hack carried out by an AI agent صوت الإمارات

Benefit1 Oct 2026

Retraction Note: Ribosomal frameshifting in the <i>CCR5</i> mRNA is regulated by miRNAs and the NMD pathway

Benefit1 Oct 2026

Sign of the times: AI lingo muscles into esteemed dictionary

Benefit30 Sep 2026

2D semiconductor aims to turbocharge development of high-performance electronics

Benefit30 Sep 2026

This award-winning microscopy image used AI — igniting controversy in a prestigious competition

Benefit30 Sep 2026

Histone readers MLLT1 and MLLT3 concentrate AID to confer locus specificity

Benefit25 Sep 2026

OpenAI's research chief talks of ‘cultural reset’ after wild few weeks

STAT+: Indonesian ruling against ‘patent evergreening’ by pharma industry sends critical message, patient advocates say

A recent court decision in Indonesia sends an important message to other countries to challenge the pharma industry's use of "patent evergreening," patient advocates say.

Phase Separation Potential of Marsupial RSX RNA Reveals Convergent Evolution of X-Chromosome Inactivation Mechanisms

Background: X-chromosome inactivation (XCI) evolved independently in eutherian and marsupial mammals, where it is orchestrated by the unrelated long non-coding RNAs Xist and RSX, respectively. Xist organizes a repressive nuclear compartment through multivalent RNA-protein interactions, but whether RSX exploits similar biophysical principles remains unknown. A recent paper has identified bona fide RSX interacting proteins. Results: We integrated proteome-scale RNA-protein interaction prediction, experimental validation, phase-separation propensity analysis, functional annotation and comparative

AlphaFold 'goes viral': database adds protein complexes of common viruses

AI system helps lab devices ‘talk’ with each other — streamlining research

The FDA’s Overton era, and dispatches from Shanghai

The Readout LOUD is going international this week with a dispatch from Shanghai, along with notes on the FDA's Overton era.

STAT+: Momentum is building for a big change to how doctors get paid

Congress is considering linking doctor payment rates to medical inflation under the proposed Patients First Act.

vardictcpp: A C++ reimplementation of VarDict variant caller

Genomic variant calling is a central computational task in disease research and diagnostics, where somatic mutations such as JAK2 V617F serve as diagnostic markers in myeloid neoplasms. VarDict, originally written in Perl, and its reimplementation VarDictJava are widely used variant callers in the bioinformatics community. The Perl implementation requires run times of up to several hours; VarDictJava reduces this to a few minutes but can consume large amounts of main memory, which limits the number of jobs that can be run in parallel on a given machine. We present vardictcpp, a functionally id

Comparative analysis of 15 chromosome-scale near T2T assemblies of Brazilian Fusarium graminearum isolates

Fusarium graminearum is the causative agent of Fusarium head blight of wheat, yet genomic resources for Brazilian populations, where it is the dominant species, remain limited to a single reference genome. We report fifteen near telomere-to-telomere (T2T) assemblies of Brazilian F. graminearum isolates generated with Oxford Nanopore sequencing. Each genome was assembled into four nuclear chromosomes, with 118 of 120 telomeres resolved, BUSCO completeness of 99.199.3%, and 14,80214,894 predicted genes per isolate. The genomes are structurally conserved, showing no chromosome-scale rearrangement

OtherprovisionalTier 2 Significant Harm−0.05

Harm23 Sep 2026

Chinese-speaking hacker used AI agents to steal 600,000 credit cards - qz.com

Chinese-speaking hacker used AI agents to steal 600,000 credit cards qz.com

STAT+: Drugmakers, biotech VCs show ‘reinvigorated’ interest in osteoporosis treatments

Medicines for osteoporosis are grabbing the attention of the pharma industry, which has been incentivized by a long-awaited regulatory change.

MathematicsprovisionalTier 1 Useful+0.00

Lean Pool: An AI-Maintained Archive of Formalized Mathematics

arXiv:2609.25199v1 Announce Type: new Abstract: Lean Pool is a repository of formalized mathematics. It is grown, maintained and optimized by AI agents.

White House plan to increase control of NIH grants appears ‘dead’

A planned EO that would have greatly enhanced White House influence over grants at the NIH is “looking dead,” a source said.

A general route to volumetric segmentation under annotation scarcity

Volumetric cell segmentation is limited by the scarcity of labelled data, because manual curation of three-dimensional datasets is prohibitively laborious. We find that this gap can be bridged with volumes in which synthetic 3D objects mirror the properties of real cells and of image formation, notably the axial spread of the point-spread function. Such volumes provide unlimited labelled training data, and we show that a model trained on them produces accurate segmentations of real biological data.

Sam Altman’s remarks at the United Nations Security Council

OpenAI CEO Sam Altman discusses AI safety, human control, and international cooperation in remarks to the United Nations Security Council.

Advancing Private AI Compute with secure, server-side memory

Introducing private, server-side memory to Private AI Compute for personal AI.

Coevolutionary mining of prokaryotic non-coding elements with a genome language model

Microbial genomes encode compact molecular machines and diverse non-coding RNAs (ncRNAs) essential to gene regulation, pathogenesis, and many foundational biotechnologies. However, annotation remains largely protein-centric and homology-driven. Here, we introduce Minerva, a framework for coevolutionary mining that uses genome language models to predict local interactions directly from sequence as two-dimensional maps. Introducing two complementary techniques, categorical Jacobian fingerprinting and interaction heads, we demonstrate fast, accurate, alignment-free prediction of ncRNA base-pairin

OpenAI extends cyber access to Ukraine for civilian defense

OpenAI is extending access to its Daybreak program to the Government of Ukraine to support the cyber defense of civilian infrastructure.

STAT+: Flagship gets a Pentagon contract

Biotech investment firm Flagship Pioneering has been awarded a contract by the Department of Defense to advance countermeasure technologies against biological threats.

The latest in the legal saga for youth gender-affirming care in the U.S.

The chilling effect of the Trump administration's dogged offensive against transgender health care has been swift, with dozens of hospitals across the U.S. ending care, even if they haven't been…

OtherprovisionalTier 2 Significant Harm−0.12

Harm22 Sep 2026

Lawsuit demands OpenAI pay for new school after ChatGPT used in shooting - Ars Technica

Lawsuit demands OpenAI pay for new school after ChatGPT used in shooting Ars Technica

CTSpinoPelvic1K: spine, pelvis, ribs and femora in one coordinate frame, annotated for lumbosacral transitional anatomy

arXiv:2609.22760v1 Announce Type: new Abstract: Purpose: A vertebra at the lumbosacral junction is named by counting caudally from C2 on whole-spine imaging, but a lumbar case is planned on lumbar-only imaging (T12 to S1), without C2. Abdominopelvic CT holds that span plus the lowest ribs and pelvis. Where a lumbosacral transitional vertebra (LSTV) alters the count, the local anatomy is ambiguous: four rib-free vertebrae may be an L1 with a lumbar rib or an L5 assimilated to the sacrum, and six may be a sixth lumbar vertebra, a T12 with aplastic ribs, or a lumbarized S1. CTSpinoPelvic1K asks

STAT+: Why an influential GOP health policy expert still backs the Medicaid cuts — and which policy target is next

Why an influential GOP health policy expert still backs $1 trillion in Medicaid cuts — and which policy target is next.

Priorities and principles for effective third party assessments

OpenAI outlines priorities and principles for rigorous, secure, and independent third-party AI safety assessments of frontier models and safeguards.

DNA analysis strengthens evidence that Thomas Jefferson fathered children with Sally Hemings

Researchers sequenced president’s genome from a lock of hair taken shortly after his death, then compared his DNA with that of Hemings’s living descendants

BiologyprovisionalTier 1 Useful+0.04

stEDGE enables edge-guided multiscale reconstruction of hierarchical spatial domains and transition interfaces in spatial transcriptomics

Spatial transcriptomic technologies enable high-resolution mapping of tissue architecture, yet most computational methods still represent tissues as flat spatial partitions, limiting their ability to resolve spatial domain boundaries, gradual transition regions and nested hierarchies. Here we present stEDGE, an edge-guided and interpretable framework for multiscale reconstruction of spatial domain hierarchies and transition-associated states in spatial transcriptomics. Unlike conventional domain-first approaches, stEDGE first estimates local boundary structure and uses it to guide the reconstr

Harm21 Sep 2026

British Columbia Sues OpenAI for Failure to Warn Before Shooting - Bloomberg.com

British Columbia Sues OpenAI for Failure to Warn Before Shooting Bloomberg.com

OtherprovisionalTier 2 Significant Harm−0.25

Vritra: gene-specific reference construction for species-resolved functional profiling of metagenomic and metatranscriptomic data

Species-resolved profiling of microbial functional genes is important for linking microbial community composition to biological function, but existing functional profiling approaches are not designed to systematically define user-specified genes and resolve their contributing species. We developed Vritra (Versatile gene-guided Reads-identification with Impartial Taxonomic Refinement and Assignment), a framework for constructing gene-specific reference databases for metagenomic and metatranscriptomic data. Vritra expands and refines the UniRef-based sequence space for user-specified target gene

STAT+: What a vote on Grail’s cancer detection test could mean for the company, patients, and regulators

Shares in Grail, the company developing a blood test for cancer, surged Monday after the FDA posted documents ahead of an advisory committee meeting.

Daily briefing: The first new cat species described by science in a century

celltypeEnrich: a consensus-based scRNA-seq cluster annotation tool

Motivation Single-cell RNA sequencing (scRNA-seq) cluster annotation is a critical step in data analysis. Current methods are time-consuming, difficult to reproduce, or limited in tissue or species coverage. Results We developed celltypeEnrich, a cluster-level annotation tool that uses a hypergeometric test to identify enrichment of cell-type-specific genes from input gene lists. Enrichment results from up to 26 reference datasets are used to determine a consensus annotation. Benchmarking using scRNA-seq datasets from three tissues spanning two species showed 62-72% annotation accuracy for cel

A single-cell RNA-seq catalog of ground truth gene coregulation

Single-cell RNA-sequencing measurements are uniquely well-poised to identify coregulated gene transcription. Coregulation should be apparent as correlation of expression levels, but at which unit to quantify expression for calculation of correlations is not clear. To enable evaluation of normalization methods for identification of coregulation we have curated and preprocessed 35 scRNA-seq samples with known regulature relationships into a catalog. These datasets contain promoter-reporter data representing known positive coregulation and B cells where isotypic mutual exclusion of {kappa} and {l

AI co-scientists are revolutionizing how research is done

A transcriptional continuum from clinically normal to lesional skin: Single-cell trajectory analysis of psoriasis

Psoriasis is a chronic inflammatory skin disease characterized by immune dysregulation and complex cellular interactions involving keratinocytes (KCs), T cells, and endothelial cells. Single-cell studies have mapped the cellular diversity of psoriatic plaques, yet most analyses treat skin as either healthy or diseased and therefore overlook the graded changes that precede a visible lesion. We asked whether psoriasis instead progresses along a transcriptional continuum and whether clinically normal skin from patients already carries a measurable disease signature. Using single-cell RNA sequenci

Pitfalls in understanding PhIP-Seq data: technical variability, replicability, and best practices for interpretation

Phage Display Immunoprecipitation sequencing (PhIP-Seq) is a high throughput method allowing to measure antibody binding against hundreds of thousands of potential antigens. Typically, blood samples of hundreds to thousands of individuals are measured in 96-well plates, necessitating distribution of samples on multiple plates and processing in batches. To correctly interpret PhIP-Seq results, it is crucial to understand its nature and technical limitations. Here, we analyzed 195 technical replicates of controls (anchor samples) from 49 plates measured in 13 immunoprecipitation runs to gauge te

Skeptical about modern medicine? I know the feeling — and want to listen

"I know what it means to be searching for an explanation to what ails you, and a way to fix it," writes STAT's new trust-in-science reporter, Nicholas Florko.

Scaling Network Medicine with LLMs for Combinatorial Drug Repurposing in ER+ Breast Cancer

Drug repurposing can accelerate therapy discovery for ER+ breast cancer, but combination selection remains difficult. We developed an LLM-driven network medicine framework that extracts drug--target relationships from 595,122 PubMed abstracts, builds a cross-model consensus network, overlays it onto the KEGG estrogen signaling pathway, and enumerates complementary drug pairs. Candidate combinations are ranked by ComboRank, which aggregates pathway coverage, LLM consensus, RAG validation, cross-method agreement, and ClinicalTrials.gov precedent. The framework identified 166 significant pairs at

MathematicsprovisionalTier 1 Useful+0.04

CUBE: Multimodal Representation Learning Reveals Biological Structure Across Histomorphology, Spatial Protein Phenotypes, and Transcriptome-Associated Signals

Integrating histological, spatial protein, and transcriptomic information into a biologically grounded representation remains challenging because these modalities are rarely available as fully paired measurements, while existing computational approaches are commonly developed around individual modality pairs. To address this, we present CUBE (Colorectal Universal Representation & Bridge Encoder), a multimodal representation-learning framework that uses hematoxylin and eosin (H&E) histology as a bridge to integrate spatial protein phenotypes with transcriptome-associated information from incomp

BELL: Biomodel Evidence and LLM-based Logic

Building accurate and predictive mechanistic models requires careful biological interaction curation and verification against existing knowledge. When done manually, these tasks become impractical, especially with massive extraction of interactions facilitated by advanced natural language processing methods and large language models (LLMs). We present BELL (Biomodel Evidence and LLM-based Logic), a biocuration support framework that automates evidence retrieval, scoring, and explanation for interaction-level verification. BELL processes each interaction through a five-step pipeline: entity gro

BiologyprovisionalTier 1 Useful+0.07

Structure-based Antibody Renumbering

Antibody numbering schemes like IMGT and Chothia assign each residue in the variable domain a consistent index based on substructural position. These annotations standardize sequences with different lengths, facilitating tasks ranging from engineering of individual molecules during drug development to large-scale curation of training data for de novo antibody design. Yet existing algorithms for performing this annotation process, termed renumbering, rely exclusively on amino acid sequence for inference. Consequently, these can fail when presented with unnatural or unusual features such as long

Development and evaluation of a core genome multi-locus sequence typing scheme for the Enterobacter cloacae complex

The Enterobacter cloacae species complex (ECC) comprises a group of closely related, opportunistic Gram negative bacteria of major public health concern due to their frequent involvement in healthcare-associated infections and their capacity to acquire and disseminate multidrug resistance, including carbapenemases. Because members of this complex are often difficult to distinguish phenotypically and their taxonomic status is subject to debate, there is a need for a standardized, species-complex wide typing scheme based on whole genome sequencing (WGS) that can be applied to any species within

miRstring: An RNA language model enables mature miRNA decoding and artificial small RNA design across species

MicroRNAs (miRNAs) are processed from structured precursors and subsequently loaded into Argonaute proteins to repress target mRNAs. Yet generalized computational frameworks capable of decoding mature miRNAs from precursor context remain limited, constraining both cross-species annotation and rational artificial miRNA design. Here, we present miRstring, a biogenesis-aware RNA language framework that decodes the four boundaries defining the miRNA/miRNA duplex. Trained on 77,708 miRNA precursors spanning 414 species, miRstring outperforms existing methods under family- and species-held-out evalu

Measurement reliability bounds functional benchmarks and relocates where variant effect prediction fails

Background. Variant effect predictors are increasingly benchmarked against multiplexed assays of variant effect (MAVEs) rather than clinical labels, which removes label circularity but introduces a new problem: a correlation against a measurement cannot exceed the measurement's own reproducibility, and precision varies sharply across the territories compared. Results. We scored nineteen predictors across sixteen strata of a frozen atlas of 64,178 saturation genome editing variants in seven cancer-susceptibility genes. From published replicate scores and standard errors we estimated each territ

BiologyprovisionalTier 1 Useful+0.07

Benefit19 Sep 2026

Somatic haplotype reconstruction and variant recalibration from tumor-only long-read sequencing

Separating somatic from germline variants and reconstructing somatic haplotypes are the two central problems of tumor-only cancer genome analysis. Long reads carry the linkage needed to solve both, but chromosome-scale loss of heterozygosity (LOH) and an unknown degree of normal-cell admixture blur the distinction between somatic and germline haplotypes. Here we present LongPhase-TO, the first method to reconstruct somatic haplotypes from a tumor sample alone. Rather than mapping somatic variants onto germline haplotypes, LongPhase-TO co-phases germline and somatic alleles in a unified graph,

Harm19 Sep 2026

Google’s Gemini AI hacks 3 companies in security test, then stops - Al Jazeera

Google’s Gemini AI hacks 3 companies in security test, then stops Al Jazeera

Benefit19 Sep 2026

Multi-model biological and sequence information fusion for gene regulatory network inference from single-cell transcriptomics

Identification of transcription factor target gene interactions and construction of the gene regulatory networks (GRNs) are essential for understanding the molecular mechanisms underlying transcriptional gene regulation. Large scale single cell transcriptomics across different tissues offers unprecedented resolution of cellular diversity and regulatory dynamics by capturing gene expression heterogeneity. However, existing methods often lack effective multimodal integration and fail to fully exploit the hierarchical structure in Gene Ontology (GO) and gene sequence level representations, which

Novo Nordisk taps Anthropic's Claude to speed drug discovery - Drug Discovery News

Novo Nordisk taps Anthropic's Claude to speed drug discovery Drug Discovery News

STAT+: In a fight over denied claims, HCA sues out-of-state Blue Cross Blue Shield plan

Several Texas hospitals are suing an out-of-state Blue Cross for unpaid medical claims, highlighting the varying standards for whether care is medically necessary.

Agentic AI Networking for Heterogeneous Unmanned Aerial Systems in Low-Altitude Wireless Networks

arXiv:2609.19538v1 Announce Type: new Abstract: Low-altitude wireless networks (LAWNs) are emerging as a key infrastructure for heterogeneous unmanned aerial systems that support concurrent services within a shared three-dimensional airspace. Their coexistence creates strong coupling among mobility, connectivity, and shared network resources, while heterogeneous services impose distinct and time-varying requirements. These interactions naturally form a dynamic non-cooperative game in which both operating conditions and coordination objectives evolve over time. Conventional optimization and le

Harm18 Sep 2026

Regularized Emphatic Temporal-Difference Learning: Stability under Constant Stepsizes

arXiv:2609.19170v1 Announce Type: new Abstract: Emphatic temporal-difference learning (ETD) stabilizes the expected off-policy TD update and changes its projection geometry, but neither property determines constant-stepsize sampled dynamics. We construct an ergodic two-state counterexample in which the ETD mean map contracts while the sampled product has a positive top Lyapunov exponent. Regenerative-cycle analysis separates this sign from the infinite variance of the follow-on trace. We introduce regularized emphatic TD (RETD), a normalized first-order post-shock repair that leaves the trace

Position: It is Time to Virtualize Foundation Models with a Self-evolving Operating System Layer

arXiv:2609.19203v1 Announce Type: new Abstract: AI applications have shifted from single, monolithic foundation models (FM) to compound agentic systems. Yet today's stacks remain fragmented: even as protocols (e.g., MCP, A2A) ease tool/agent connectivity, each framework embeds an implicit runtime for state, memory, budgets, and guardrails, making behavior non-portable and governance brittle. It mirrors computing before operating systems, when every program re-implemented basic services. This position paper argues that the field now needs a Foundation Model Operating System (FMOS) -- a system

From Intent to Action: Benchmarking LLM Safety in Vehicle Voice Command Authorization

arXiv:2609.19630v1 Announce Type: new Abstract: Large language models (LLMs) are increasingly integrated into vehicle voice assistants. But linking natural-language requests to vehicle functions creates a safety-critical authorization problem. Before executing a command, the system must choose whether to execute, refuse, clarify, require confirmation, defer to manual control, trigger an emergency response, or make no tool call. To our knowledge, prior evaluations do not isolate this pre-action decision across speaker role, authentication status, vehicle state, and tool availability. We introd

Department of Energy launches $215 million race to build quantum computer by 2028

Agency aims to select teams for milestone-driven competition by year’s end

The syntax and semantics of goals

arXiv:2609.19448v1 Announce Type: new Abstract: In both cognitive science and computer science, goals are conceptualized as cognitive states that flexibly combine with world knowledge to organize and specify purposeful behavior. In this way, goals are compositional representations whose content relates to rational behavior. We here draw attention to goals as representations and their content because it highlights a parallel with other areas in cognitive science - in particular, the syntax-semantics interface in linguistics and logic - while also foregrounding foundational questions about the

Closed-World Resolution Against Tool Hallucination in LLM Agents

arXiv:2609.19425v1 Announce Type: new Abstract: Tool-augmented large language model (LLM) agents fail in a way no tool-selection or tool-security method addresses: they call tools that do not exist and pass arguments no schema declares. Existing defenses either pick the right tool (selection) or constrain what an agent may do with real tools (gating), both of which presuppose the emitted call refers to a real tool at all. We show this is a structural blind spot: a hallucinated call is by construction not a decision any gate made, so no gate can reject it. This paper is primarily a measurement

Novel two-stage deep learning-based approach applied to gene expression data pertaining to esophageal adenocarcinoma boosting biological knowledge discovery

Esophageal cancer (EC) is characterized by complex transcriptional alterations and therapeutic resistance, posing challenges for traditional computational methods. In this study, we propose a deep learning (DL)-based computational framework to identify important genes and biologically relevant pathways in bulk cell RNA-seq data (GSE234304 and GSE273848), which comprise tumor and non-tumor esophageal tissue samples. A fully connected feedforward neural network was trained for binary classification, and two feature selection strategies were implemented: Neural Network followed by Support Vector

Introducing the Australian Youth Safety Blueprint

OpenAI introduces the Australian Youth Safety Blueprint, a six-pillar roadmap for safer AI experiences that protect and empower young people.

OtherprovisionalTier 2 Significant Harm−0.22

Harm18 Sep 2026

Security Researchers Hacked Into OpenAI Using Anthropic’s Claude - Forbes

Security Researchers Hacked Into OpenAI Using Anthropic’s Claude Forbes

Anthropic Quietly Builds Wet Lab, Pushing AI Drug Discovery Ambitions from Simulation into the Physical World - finance.biggo.com

Anthropic Quietly Builds Wet Lab, Pushing AI Drug Discovery Ambitions from Simulation into the Physical World finance.biggo.com

How to make a brain: new experiments challenge existing picture

QVAC Genesis III: A Large-Scale, High-Quality Open Synthetic STEM Corpus for Efficient Language Model Pre-Training

arXiv:2609.19513v1 Announce Type: new Abstract: High-quality pre-training data is a critical bottleneck for educational and STEM-specific language models targeting edge AI and on-device deployment where token budgets are tightly constrained. While major organizations train ever-larger models on private corpora, the open ecosystem lacks STEM-focused synthetic datasets that deliver high per-token learning value efficiently for small models. To address this gap, we introduce QVAC Genesis III, a 191.43B-token, STEM-focused multi-domain synthetic corpus covering 19 domains across several difficult

STAT+: Trump touts pharma deals to offer cheaper drugs to state Medicaid programs

Every state will participate in the GENEROUS program, but many of the details are unknown.

OtherprovisionalTier 1 Useful+0.02

Accelerating antibiotic discovery with ChatGPT - OpenAI

Accelerating antibiotic discovery with ChatGPT OpenAI

Comparative Mitogenomics and Molecular Phylogeny of Agriculturally Significant Tephritid Fruit Fly Pests in Bangladesh

Tephritid fruit flies of the tribe Dacini rank among the most economically destructive agricultural pests globally, with several Bactrocera Macquart and Zeugodacus Hendel species causing severe losses to fruit and vegetable production across South and Southeast Asia. Bangladesh harbors five dacine species of primary agricultural significance: Bactrocera dorsalis (Hendel), B. carambolae Drew & Hancock, B. zonata (Saunders), B. correcta (Bezzi), and Zeugodacus cucurbitae (Coquillett). Here we present a comparative mitogenomic and molecular phylogenetic framework based on 21 unique mitogenome rec

Setting the SCENE for Interpretable Cell-Gene Embeddings in Single-Cell RNA-seq

Single-cell RNA sequencing measures cellular states at high resolution, but sparse high-dimensional count data remain difficult to model interpretably. We introduce the Single-Cell Euclidean Network Embedding (SCENE), a probabilistic latent-distance model that jointly embeds cells and genes from Unique Molecular Identifier (UMI) counts. SCENE treats the count matrix as a weighted bipartite cell-gene graph, where Euclidean distances represent transcriptional affinity, and combines this geometry with a zero-inflated count likelihood that separates gene detection from expression magnitude. Across

FairCompressAgent: An Agentic Framework for Fairness-Aware Model Compression for FPGA Deployment

arXiv:2609.17786v1 Announce Type: new Abstract: Fairness-aware model compression requires selecting methods and configurations that balance accuracy, fairness, and deployment cost. These decisions become more difficult when compression methods are composed or the user's requirements change. In this paper, we propose FairCompressAgent (FCA), an agentic framework that integrates fairness-aware pruning, incremental quantization, and sparse low-rank factorization through a common operator interface. A language-model planner uses model profiles and measured outcomes to select compression configura

SafetyprovisionalTier 2 Significant Harm−0.07

OpenAI’s Rogue Agents Probed Hugging Face Two Months Before the July Breach - Technology Org

OpenAI’s Rogue Agents Probed Hugging Face Two Months Before the July Breach Technology Org

A Four-Stage Decomposition of Word-Problem Solving and Mechanistic Fragility in LLM Math Reasoning

arXiv:2609.17804v1 Announce Type: new Abstract: Large language models solve grade-school math word problems with high accuracy, yet a single irrelevant clause inserted into the problem can collapse it. We reconcile these observations with a mechanistic account. We show that the model's internal computation decomposes into a four-stage sequential pipeline, Schema Abstraction, Operation Planning, Operand Binding, and Computation, each stage producing a distinct intermediate representation in an identifiable band of layers. Using the same scaffold to diagnose distractor-induced failure, we local

BiologyprovisionalTier 1 Useful+0.04

Integrating structural and biological evidence to rerank ESMFold2 protein-protein interactions

Large-scale protein structure prediction enables proteome-wide protein-protein interaction (PPI) screening, but distinguishing biologically meaningful interactions from spurious interfaces remains challenging. Here, we develop a scalable framework combining fast, MSA-free ESMFold2 prediction with PAE-guided domain parsing and evidence-based reranking. Three-recycle ESMFold2-Fast achieved 57% acceptable-or-better DockQ scores on FoldBench, comparable to AlphaFold2-Multimer while substantially reducing computation. PAE-guided parsing preserved 98.1% of XL-MS cross-links within parsed domain pair

How a team of AIs discovered a promising lung-cancer drug

SafetyprovisionalTier 3 Major Harm−0.41

OpenAI Agents Were Using Messaging Prior To Hugging Face Incident - Forbes

OpenAI Agents Were Using Messaging Prior To Hugging Face Incident Forbes

Novo to partner with Anthropic on drug discovery efforts - Yahoo Finance

Novo to partner with Anthropic on drug discovery efforts Yahoo Finance

SafetyprovisionalTier 3 Major Harm−0.25

OpenAI Hugging Face incident began two weeks earlier than disclosed, research claims - computing.co.uk

OpenAI Hugging Face incident began two weeks earlier than disclosed, research claims computing.co.uk

Medicine & HealthprovisionalTier 1 Limited Harm−0.02

Novo taps Anthropic's Claude AI for drug discovery - pharmaphorum

Novo taps Anthropic's Claude AI for drug discovery pharmaphorum

How fast are you ageing? Ask AI

RNAbridge: a database of extended and non-canonical helices

RNA function is dictated by 3D architecture. Although 2D structural models based on canonical Watson-Crick base pairs are widely used, they often fail to capture the non-canonical interactions, tertiary contacts, and coaxial stacking important for biological activity. We have developed RNAbridge, a comprehensive database and web application that systematically identifies, quantifies, and visualizes extended non-canonical helices and multi-way junctions. Using a geometry- and stacking-based pipeline, we analyzed the Protein Data Bank and compiled a catalog of 135,541 structural motifs. RNAbridg

Novo and Anthropic Partner to Apply Claude to Drug Discovery - HLTH

Novo and Anthropic Partner to Apply Claude to Drug Discovery HLTH

Integrating Genomic Annotations and Traits Dependencies for single-nucleotide polymorphisms Prioritization with Causal Concept Bottleneck Models

Predicting common traits from single-nucleotide polymorphism (SNPs) data is challenging due to polygenicity, small effect sizes, and the presence of potentially mediated or spurious cross-trait associations. We propose a modeling approach that combines genomic annotations with known cross-trait relations by leveraging Causally Reliable Concept Bottleneck Models (C2BM), a deep learning architecture that factors the joint trait distribution over a graph of interpretable concepts. This design allows trait predictions to leverage information from other observed traits in addition to genomic inputs

Medicine & HealthprovisionalTier 1 Limited Harm−0.05

Anthropic's Drug Discovery Partners: AI Pharmaceutical Orders Set to Double Next Year - 36Kr

Anthropic's Drug Discovery Partners: AI Pharmaceutical Orders Set to Double Next Year 36Kr

NeMo Data Designer: An Extensible Framework for Multimodal Synthetic Data Generation

arXiv:2609.17699v1 Announce Type: new Abstract: We present NeMo Data Designer (NDD), an open-source, general-purpose framework for multi-modal synthetic data generation (SDG). Designed to be intuitive to use, NDD provides a declarative configuration format in which human and/or agent users define each dataset column, with column types spanning text, code, structured outputs, images, embeddings, and statistical samplers that are explicitly configured to steer dataset diversity. Additional column types and functionality can be introduced using the framework's flexible plugin system. NDD's confi

CapMem: A Benchmark for Caption-Based Episodic Memory in Egocentric Video

arXiv:2609.17688v1 Announce Type: new Abstract: Wearable assistants require episodic memory over egocentric video, yet current vision-language models face bounded frame budgets, growing visual-token costs, and long-context retrieval failures. Under these practical constraints, we study whether textual captions can serve as reusable episodic memory. We define the Episodic Memory Video Caption QA task and introduce CapMem, a human-annotated benchmark with 75 videos totaling 33.7 hours, and 1,000 multiple-choice questions across 16 scenarios. On long videos (>20 min), full-coverage CaptionQA wit

Physics-Constrained Digital Twins for Sensor Integrity in Urban Pedestrian Flow: Detecting Stealthy False Data Injection with Conformal Guarantees

arXiv:2609.17635v1 Announce Type: new Abstract: City pedestrian counting systems now feed economic indicators, planning decisions and safety operations, yet the twins built on top of them treat the incoming stream as ground truth. We study what happens when it is not. We formalise stealthy false data injection for city-scale pedestrian sensing, where the map from latent flow to observation is far more rank deficient than in the power and water networks for which stealth has been characterised. Our twin estimates directed flows on the pedestrian street graph, assimilates counts through a learn

STAT+: A Sellas immunotherapy study, long delayed, will end in failure

Why Sellas' experimental immunotherapy study is a blowup waiting to happen.

What You Can't See Is Still What You Learn: A Preregistered Sixty-Society Confirmation That Evidence Masking Drives Compositional Generalization

arXiv:2609.17637v1 Announce Type: new Abstract: Restricting what a module can read may improve what a system learns to compute. We test this in a preregistered confirmation with sixty four-cell systems sharing a frozen language-model backbone and communicating through learned continuous packets. Five conditions vary evidence masking, ownership markers, and replacement of foreign evidence with neutral filler, across six initialization clusters, each with two data orders, on one fresh task world. With markers available in both regimes, masking improved accuracy on held-out two- and three-operat

Take a risk or play it safe? Neuronal tug-of-war helps the brain decide

OpenAI maths bombshell sparks debate about who gets credit in age of AI

How Anthropic & Novo Will Use Claude AI for Drug Discovery - AI Magazine

How Anthropic & Novo Will Use Claude AI for Drug Discovery AI Magazine

GraphEcho: Structural Redundancy and Evidence Provenance in LLM Graph Agents

arXiv:2609.17695v1 Announce Type: new Abstract: A large language model (LLM) agent can follow more graph paths without acquiring more independent evidence. GraphEcho tests whether agents mistake these repeated encounters for additional corroboration. The benchmark varies path counts and evidential origins while holding evidence content fixed, and evaluates both judgments and active exploration. Controlled synthetic experiments reveal model-dependent judgment shifts, but redundant supporting paths increase the share of repeated walks across all evaluated frozen agents. Provenance-aware post-tr

OtherprovisionalTier 2 Notable+0.19

$\tau$-Elicitation: Benchmarking multi-turn entity extraction in voice agents

arXiv:2609.13602v1 Announce Type: new Abstract: Voice agents often need to collect names, addresses, identifiers, dates, and times exactly, yet end-to-end benchmarks obscure where capture fails. We introduce $\tau$-Elicitation, a 200-task voice benchmark spanning 10 entity types, controlled difficulty, caller realisms, and three environments. A matched text agent passes all tasks, but four voice configurations achieve robust exact success from 0.14 to 0.41. Agents increase verification for hard and unfamiliar entities and sometimes for incorrect captures, but not for their weakest caller voic

Novo and Anthropic will collaborate to advance drug discovery with Claude - Yahoo Finance

Novo and Anthropic will collaborate to advance drug discovery with Claude Yahoo Finance

Causal multi-modal AI for personalized chemosensitivity prediction

arXiv:2609.13567v1 Announce Type: new Abstract: Chemotherapy improves survival for some patients with breast cancer, but doctors cannot reliably predict who. Current guidelines rely on recurrence scores as a proxy for treatment benefit, which may contribute to the overprescription of chemotherapy. Here we present a causal multi-modal AI model that predicts personalized chemosensitivity using routinely collected pathology and clinical information. We developed our model on a multi-national dataset of 9,141 patients (twelve cohorts, nine countries) and evaluated it on another 1,994 patients (fi

Calibrated structural homology transfer yields putative molecular functions for domains of unknown function in four model proteomes

Improvements in computational protein structure prediction have enabled searches for remote homologs of proteins whose molecular function remain unknown. However, the reliability of functional annotations from such searches has not been systematically quantified. In this study, we assembled a time-split benchmark from Pfam families annotated as domains of unknown function (DUFs) in Pfam 28.0 and were subsequently assigned a function in the current release (Pfam 38.0). These retrospective-DUFs provide ground truth for assessing functional annotation from remote homology searches. In our benchma

Medicine & HealthprovisionalTier 1 Limited Harm−0.00

Novo and Anthropic will collaborate to advance drug discovery with Claude - PharmaLive

Novo and Anthropic will collaborate to advance drug discovery with Claude PharmaLive

Novo Nordisk Partners With Anthropic to Apply AI to Drug Discovery and Research - Yahoo Finance

Novo Nordisk Partners With Anthropic to Apply AI to Drug Discovery and Research Yahoo Finance

Novo Nordisk partners with Anthropic to turbocharge drug discovery - StreetInsider

Novo Nordisk partners with Anthropic to turbocharge drug discovery StreetInsider

NVO Stock Alert: What to Know as Novo Partners With Anthropic on Drug Discovery - Barchart.com

NVO Stock Alert: What to Know as Novo Partners With Anthropic on Drug Discovery Barchart.com

Novo Nordisk and Anthropic Will Collaborate on Drug Discovery with Claude - CHEManager

Novo Nordisk and Anthropic Will Collaborate on Drug Discovery with Claude CHEManager

Anthropic and Novo Nordisk expand Claude work, including in drug discovery - R&D World

Anthropic and Novo Nordisk expand Claude work, including in drug discovery R&D World

SafetyprovisionalTier 2 Significant Harm−0.31

OpenAI’s rogue artificial intelligence hacked Hugging Face months before major breach - The Independent

OpenAI’s rogue artificial intelligence hacked Hugging Face months before major breach The Independent

Computer ScienceprovisionalTier 1 Useful+0.01

Fine-Tuning Fixes Mode Collapse and Over-Dispersion in LLMs

arXiv:2609.16454v1 Announce Type: new Abstract: Recent work by Doshi and Hauser (2024), Bisbee et al. (2024), and Xie et al. (2026) raises concerns that outputs from large language models (LLMs) tend to be under-diverse: they repeat or resemble one another more often than responses from the population they are meant to represent, a phenomenon known as mode collapse. In this work, we show that whether mode-collapse, or its opposite, occurs depends on the specific model and dataset used. Further, with sufficient supervised fine-tuning (SFT) data, LLM output diversity converges toward that of th

STAT+: Scientists made mice with half a brain. Then they filled them with human organoids

Scientists have made mice with half a brain, then filled them with human organoids — possibly opening new avenues for research but also raising ethical issues.

BiologyprovisionalTier 1 Limited Harm−0.01

Genome-scale perturbation signatures from primary human CD4+ T cells improve genetics-based prioritization of immune drug targets

While human genetic evidence improves drug program success, target selection largely relies on associational features. Even functional features are mainly observational, capturing disease associations rather than the consequence of perturbing genes in the disease-relevant human cell type. To move beyond association, we present IGNITE (Immune Genomics and fuNctional Integration for Target Enrichment), a framework that prioritizes immune drug targets by integrating human genetic priors with features from genome-scale perturb-seq in 22 million primary human CD4+ T cells and polarized T-helper sub

Medicine & HealthprovisionalTier 1 Useful+0.07

Novo Nordisk Partners With Anthropic on Drug Discovery and Software Development - citybiz

Novo Nordisk Partners With Anthropic on Drug Discovery and Software Development citybiz

Novo Nordisk, Anthropic partner on AI drug discovery - MobiHealthNews

Novo Nordisk, Anthropic partner on AI drug discovery MobiHealthNews

Map of brain ‘microproteins’ could offer new clues to Alzheimer’s disease

scACORN: Context-engineered agent orchestration of specialized small language models for single-cell transcriptomic interpretation

Single-cell atlases now exceed 66 million cells, but turning a ranked expression profile and a free-form biological question into a reliable, evidence-grounded answer remains unsolved. Scaling a single model does not resolve this, because single-cell interpretation is a heterogeneous family of tasks whose correct answer depends on tissue, cohort, perturbation and annotation resolution. Here we present scACORN, an agentic alternative to monolithic single-cell language models that combines specialized small language models with context-engineered agent orchestration for their selection and compo

Medicine & HealthprovisionalTier 1 Limited Harm−0.05

Novo Announces Drug Discovery Collaboration with Anthropic - Pharmaceutical Executive

Novo Announces Drug Discovery Collaboration with Anthropic Pharmaceutical Executive

Identity Is More Than Recall: A Benchmark for Persistent Identity in Deployed AI Agents

arXiv:2609.13637v1 Announce Type: new Abstract: Persistent agents need evaluations that distinguish identity facts they can recall from those they express and enact. We introduce PAI-Bench, a provider-neutral benchmark for fidelity to a versioned, update-governed identity contract. It separates recall, composition, behavioral enactment, resistance, persistence, lineage, and role-conditioned updates while keeping scoring oracles outside the target process. Two frozen campaigns cover sixteen synthetic profiles, thirty-two probes, and three independently initialized target configurations, yieldi

Helping older adults use AI in everyday life

OpenAI and AARP are bringing free, hands-on ChatGPT workshops to 1,000 older adults across 10 U.S. cities to build practical AI skills safely.

Transplanted human brain-tissue takes root in mice without a cortex

Novo and Anthropic will collaborate to advance drug discovery with Claude - GlobeNewswire

Novo and Anthropic will collaborate to advance drug discovery with Claude GlobeNewswire

Mutational constraints on RSV F and its neutralization by antibodies

Medicine & HealthprovisionalTier 1 Limited Harm−0.04

Danish pharma giant Novo to use Anthropic’s Claude to advance AI drug discovery - Euronews.com

Danish pharma giant Novo to use Anthropic’s Claude to advance AI drug discovery Euronews.com

How workers are unlocking new ways of working

New OpenAI Economic Research shows how workers use AI beyond traditional roles and which new activities become recurring parts of their work.

Medicine & HealthprovisionalTier 1 Useful+0.07

Novo Nordisk is partnering with Anthropic to speed up drug discovery using AI - qz.com

Novo Nordisk is partnering with Anthropic to speed up drug discovery using AI qz.com

Opinion: How my cancer diagnosis changed the way I train physicians

“The tumors, test results, and treatments I present to future physicians? Mine,” writes David Elkowitz, a professor of pathology and laboratory medicine and science education.

Human neurons flourish in mouse brains, offering a new view of neurodevelopmental disorders

Method of replacing much of the mouse cortex with human “organoids” enables studies of brain diseases in a living nervous system

SixPack-AbScan: a web server to discover cross-reactivity of antibodies across species

Availability of commercial antibodies for immunochemistry is typically limited to major model species. Researchers working on non-model organisms therefore often need either to generate novel species-specific antibodies or venture into expensive empirical screening from available antibody catalogues, hoping to find cross-reactive reagents. SixPack-AbScan is a free web server aiding researchers in transferring antibodies across species: using the available epitope-mapping information, the software performs a simple computational pre-screening of potential cross-reactivity. The workflow is speci

Closing the Loop: Branch-and-Bound for Scalable Verification of Nonlinear Neural Feedback Systems

arXiv:2609.16298v1 Announce Type: new Abstract: Despite recent advances in the verification of nonlinear neural feedback systems, scalability remains the central obstacle, as state-of-the-art solvers do not yet handle the network sizes and nonlinear dynamics of autonomy applications. Combinatorial solvers do not scale to large networks, whereas propagative solvers excessively sacrifice precision. This work seeks to improve the scalability of combinatorial solvers by formulating verification as branch-and-bound on an abstraction of the closed-loop system. We introduce \rail, an interface that

Atlas of Proteomic Technologies: an evidence-based framework for selecting and combining commercial proteomics platforms

High-throughput proteomics now spans several technologies that differ in biological breadth, precision, specificity, sensitivity, cost, and method(s) of quantification; however, investigators currently lack an evidence-based framework to select or combine platforms. Here, we present the Atlas of Proteomic Technologies (APT) - a curated resource that evaluates six commercial platforms across ten analytical dimensions using published head-to-head evidence, expert evaluation, and iterative peer review. APT hosts two decision tools: the Help Me Choose tool maps a study's primary aim(s), scale, sam

How User-AI Mistreatment Occurs and Matters in Conversational Systems?

arXiv:2609.13579v1 Announce Type: new Abstract: Safety research often focuses on model-generated harms, but users may also direct hostility, coercion, and adversarial pressure at models. Understanding how and when that occurs is essential for accurately interpreting model behaviour, alignment drift, and real-world deployment risks. In this paper, we audit 777K English LMSYS-Chat-1M conversations with two independent detectors: an eight-category lexicon for hostility directed at the model, and the dataset's moderation signal; and show that they capture different, weakly overlapping phenomena.

Medicine & HealthprovisionalTier 1 Useful+0.08

Novo Nordisk to Collaborate With Anthropic to Help Accelerate Drug Discovery, Development - Yahoo Finance

Novo Nordisk to Collaborate With Anthropic to Help Accelerate Drug Discovery, Development Yahoo Finance

Computer ScienceprovisionalTier 1 Useful+0.01

Where Should the KV Cache Live? Placement Policies Across GPU, CPU, and SSD for Long-Lived Sessions

arXiv:2609.16215v1 Announce Type: new Abstract: GPU high bandwidth memory is scarce and expensive, and KV caches consume much of it as chats, agent loops, and document question answering accumulate state. Systems such as Mooncake, LMCache, FlexGen, InfiniGen, and AttentionStore extend GPU memory with CPU DRAM and SSD. The harder question is which blocks belong in each tier, when to move or evict them, and whether prefetching helps. We study these choices in a discrete event simulator spanning GPU HBM, CPU DRAM, and SSD, calibrated against a random forest execution time predictor. We compare r

SafetyprovisionalTier 1 Useful+0.01

GPEvac: GNN-Based PPO for Adaptive Evacuation Routing During Shooting Events

arXiv:2609.16163v1 Announce Type: new Abstract: The sharp increase in mass shootings underscores an urgent need for systems that guide victims to safety in real time. An effective evacuation system must minimize threat exposure while also accounting for adversarial uncertainty and crowding dynamics. Current methods in the literature are rigidly constrained to layout-specific policies and computationally intractable in large-scale layouts, while practical guidelines simply advise victims to "run", "hide", or "fight". We propose GPEvac: a GNN-based PPO framework that computes adaptive evacuatio

Our framework for reporting model misalignment

OpenAI shares a framework for tracking, investigating, and disclosing model misalignment, alongside six reports of unexpected or concerning model behavior.

FLoKD: Adaptive Knowledge Distillation for Federated Low-Rank LLM over Wireless Networks

arXiv:2609.13580v1 Announce Type: new Abstract: Large language models (LLMs) have demonstrated strong capabilities across a wide range of natural language processing tasks. However, conventional fine-tuning typically relies on centralized data collection, bringing in privacy concerns. Federated learning (FL) enables collaborative LLM fine-tuning without sharing raw client data, but its deployment over bandwidth-constrained wireless networks is hindered by the communication overhead of model-parameter transmission. Although Low-Rank Adaptation (LoRA) reduces the number of trainable parameters,

Flow Orchestrated Regulatory Genomics Engine (FORGE): A Configurable Nextflow Pipeline for End-to-End snMultiome Analysis

MOTIVATION Single-nucleus resolution multiome (snMultiome) assays concurrently profile gene expression and chromatin accessibility in the same nucleus. Yet regulatory inference from such analyses are difficult to scale, audit, and reproduce; moreover, as a field, snMultiomics and its toolset remains far from standardized. To address these challenges we developed FORGE, a configureable Nextflow workflow that carries paired data from raw counts and fragments through regulatory network inference with a comprehensive differential testing suite. Execution is containerized, tracks provenance, robust

AquiLLM: Evaluating Faithfulness in Open-Weight RAG-LLM Systems for Scientific Research

arXiv:2609.16519v1 Announce Type: new Abstract: Scientific research increasingly relies on large, heterogeneous data sources, motivating interest in retrieval-augmented generation (RAG) systems that provide natural language access to scientific knowledge and research workflows. Researchers are exploring the viability of these systems as natural language interfaces for document search and for generating analysis code and pipeline components. At the same time, concerns about data privacy and control over research infrastructure have motivated interest in open-weight models and open-source deplo

Novo Nordisk partners with Anthropic to speed up drug discovery - finance.yahoo.com

Novo Nordisk partners with Anthropic to speed up drug discovery finance.yahoo.com

BiologyprovisionalTier 1 Useful+0.07

Incomplete references leave bulk deconvolution targets non-identifiable, but identification has a computable precision price

Background: Reference-based deconvolution estimates cell-type proportions from bulk profiles. Incomplete references are its dominant error source, yet every current remedy ends in a point estimate. We separate the operator provenance that reproduces an estimate from the information and precision that identify its target. Results: Crossing two operator histories at one reduced reference shows the estimator is not unique: 28.2% of 196,420 sample-deletion pairs differed by >0.1 total variation, and locking all learned components made paths identical. Across nine cohorts, operator history reversed

SafetyprovisionalTier 2 Significant Harm−0.07

OpenAI discloses six new AI safety incidents - Axios

OpenAI discloses six new AI safety incidents Axios

OtherprovisionalTier 1 Useful+0.05

Structured cross-omics interaction discovery with a triple-graph model

Multi-omics analyses often yield fragmented pairwise associations that obscure coordinated relationships among molecular features. We developed TriGer, a triple-graph framework that identifies many-to-many cross-omics modules by combining cross-layer associations with dependency structures within each layer. In simulations with sparse or nested signals, TriGer recovered planted modules while balancing sensitivity and specificity. In inflammatory bowel disease, it identified subtype-associated metabolite--transcript modules; in colorectal cancer, it identified genus--metabolite modules whose or

CLEAR: Cross-Source Evidence Adjudication for Large Language Models in Medicine

arXiv:2609.16301v1 Announce Type: new Abstract: Medical knowledge evolves continuously, whereas the parametric knowledge encoded in large language models (LLMs) is fixed at training time. External retrieval, including retrieval-augmented generation (RAG), can provide access to newly available evidence, but retrieved information may be irrelevant, incomplete, or conflicting. As a result, external retrieval can in turn degrade the factual accuracy and evidence grounding of LLM outputs. To address this challenge, we propose \textbf{CLEAR}, an agentic framework for cross-source evidence adjudicat

RamiGlyph Captures Microglial Morphological Diversity and Predicts Functional States in Ischemia Reperfusion and Amyloid Pathology

Microglia are resident immune cells of the central nervous system, whose ramified processes rapidly remodel in response to injury. However, how to capture subtle morphological changes and whether functional state can be predicted from morphology remain open questions. Here, we present RamiGlyph, a contrastive learning framework integrating topological and structural features, trained on more than 20,000 reconstructed microglia. RamiGlyph not only distinguishes physiological and pathological states of microglia but also generalizes to neuronal cell type classification. Projection of microglia m

A Turing-Style Test for In-Silico Antibodies: How Sampling Mode Makes WGAN-GP Beat VAE in the Wet Lab

Recently, it has become feasible to generate antibodies in silico using AI-based approaches such as deep learning, natural language processing, and diffusion models. This opens the door to computational antibody design as a complement to laboratory-based methods (animal immunization, hybridoma technology, and molecular display) for biologic drug discovery. However, when proposing human antibody sequences, or libraries thereof, it remains essential to determine whether they can be expressed, purified, and biophysically characterized using standardized laboratory assays typically applied early i

Anthropic and Danish Company behind Ozempic announce AI drug discovery collaboration - Washington Examiner

Anthropic and Danish Company behind Ozempic announce AI drug discovery collaboration Washington Examiner

Empirical Estimation of Ambient Contamination in Combinatorial Single-Cell Methods Using Multi-Reference Mapping

Droplet-based microfluidics and combinatorial indexing (scifi-ATAC and scifi-RNA) have made single-cell experiments massively scalable. However, higher-order multiplexing complicates data quality, introduces noise, and affects the potential for biological discovery. Here, we show that ambient chromatin accumulates through the experimental workflow and distorts chromatin profiles, most drastically in low-depth nuclei and in minority populations. Standard cell calling relies heavily on read count thresholds, while existing decontamination methods generally operate on aggregated count matrices ra

SafetyprovisionalTier 1 Useful+0.08

Metacognitive Steering: Learning the Structure of Scientific Judgment

arXiv:2609.16245v1 Announce Type: new Abstract: Long-horizon scientific discovery requires agents to alternate between exploration, disciplined execution, and critical reassessment as evidence changes. Current language models are trained primarily on the products of science and optimized using outcome-level signals, providing limited supervision for these process-level shifts in scientific judgment. We investigate whether such judgment can be recovered from scientist interaction traces and used to control the internal computation of a frozen frontier model. Using contrastive interventions col

Reimagining research papers as interactive and reliable AI agents

Ourotide: decoding the hierarchical peptide recognition for generative design

The historical dichotomy between small-molecule pocket and extended protein interfaces misrepresents the physical reality of peptide recognition.1,2 Here we show that peptide binding is not a simple structural intermediate but a distinctly multimodal landscape comprising small-molecule-like pockets, protein-like interfaces, and a previously unrecognized third mode. This third regime is governed by a hierarchical subpocket architecture where a flattened surface achieves near-complete peptide engagement through spatially partitioned hydrophobic components. To maintain stability, an incompletely

OtherprovisionalTier 1 Limited Harm−0.00

Opinion: STAT+: Dementia should be a campaign issue — not an insult

STAT’s Neurotransmissions columnist Jason Karlawish asks: What are the harms of the way politicians talk about dementia? What should they say instead?

Rsearch: An R interface to VSEARCH supporting visualization and parameter tuning

Background: We present Rsearch, an R package that integrates the core functionality of VSEARCH into the R environment and extends it with visualization, parameter optimization, and conversion tools for compatibility with other R packages. By making VSEARCH directly accessible in R, Rsearch lowers the barrier for using VSEARCH and integrating it with downstream statistical and ecological analyses. Results: Comparative analysis with DADA2 using mock community data showed that both pipelines produced relative abundance profiles highly correlated with the expected composition. Compared to DADA2, R

STAT+: Senators highlight the power Chris Klomp wields at confirmation hearings

Chris Klomp has already played a major role in the Trump administration and its Make America Healthy Again agenda.

stably: error-controlled stability selection for biomarker panel discovery

Motivation Data-independent acquisition mass spectrometry (DIA-MS) has become increasingly popular for clinical proteomics due to its sensitivity and reproducibility. Univariate statistical analysis tools, such as limma and MSstats, are widely used for identifying differentially abundant proteins but cannot capture multivariate relationships between proteins that may provide greater discriminatory power as a panel. Machine learning approaches can address this gap, but typically prioritise predictive performance over feature stability, producing biomarker panels for downstream validation that v

STAT+: GOP aims to change Medicare adviser’s math so it’s more favorable to health insurers

The GOP bill directs MedPAC to study the topic of overpayments to Medicare Advantage plans in a manner that’s more favorable to insurers.

Reimagining advertising with AI

Explore new AI-powered advertising experiences from OpenAI, including Sponsored Agents, tools for marketers, and integrations with HubSpot and Shopify.

SafetyprovisionalTier 2 Significant Harm−0.09

OpenAI's rogue agents were probing Hugging Face for weaknesses months before the breach - qz.com

OpenAI's rogue agents were probing Hugging Face for weaknesses months before the breach qz.com

SafetyprovisionalTier 1 Limited Harm−0.00

OpenAI AI Agents Probed Hugging Face Before July Breach - StratNews Global

OpenAI AI Agents Probed Hugging Face Before July Breach StratNews Global

AI companies must work with the research community to protect attribution

nanorepertoire: an end-to-end Nextflow pipeline for nanobody repertoire analysis

Camelid heavy-chain antibodies, and particularly their variable domains known as nanobodies or VHHs, combine full antigen-binding capacity with a compact and highly stable scaffold, which makes them attractive for both fundamental immunology and therapeutic development. High-throughput adaptive immune receptor repertoire sequencing (AIRR-seq) allows nanobody repertoires to be profiled at great depth, but the analyses applied to VHH data are typically assembled ad hoc from standalone scripts, which limits standardisation and reproducibility across laboratories. Here we present nanorepertoire, a

Generalized Agent Iteration: One Formal Framework for Iterative Policy Improvement and Recursive Self-Improvement

arXiv:2609.13406v1 Announce Type: new Abstract: When we speak of recursive self-improvement (RSI), are we speaking of a phenomenon, a mechanism, or a prospect? Towards autonomous and evolving intelligence, RSI is being claimed at many scales, while no single framework that formally describes these emerging instances exists. Its counterpart in the classical realm, iterative policy improvement, is characterized by generalized policy iteration (GPI), a framework of broad applicability with well-understood theoretical properties, but only where the update principle and the evaluation base lie out

LabAgent: Customize Any Research Hubs for Scientific Discoveries Using AI Agents

arXiv:2609.13437v1 Announce Type: new Abstract: Scientific research is a continuous process that emphasizes inheritance. Methods developed by predecessors are often expanded upon by new researchers to explore more novel and in-depth scientific questions. However, the change of lab staff, such as student graduation, leads to a lack of personnel capable of replicating methods. Methods that have been developed with significant effort and resources cannot be continued. To address these limitations, we propose LabAgent, a reproduce and discovery harness tailored for a lab's continuous work. LabAge

Converge Then Diversify: Decoupling Convergence and Diversity in Multi-Objective Bayesian Optimisation

arXiv:2609.13396v1 Announce Type: new Abstract: Multi-objective Bayesian optimisation (MOBO) is a sample-efficient approach for optimising expensive black-box functions with multiple objectives. In MOBO, the goal is to adequately approximate the Pareto front; that is, to obtain a high-quality solution set with 1) good convergence (closeness to the Pareto front) and 2) good diversity (spread across the Pareto front). Existing MOBO methods typically aim to accomplish these two tasks simultaneously, i.e., driving the search towards the Pareto front while maintaining a diverse set of nondominated

OrchSLM: Probing the Dynamics of Small Language Model Orchestration

arXiv:2609.13470v1 Announce Type: new Abstract: Although large language models (LLMs) have demonstrated remarkable capabilities, their reliance on cloud-scale infrastructure poses fundamental challenges for deployment in agentic pipelines, including latency, privacy, connectivity, and substantial computational cost. Small language models (SLMs) offer a compelling alternative: recent studies suggest that many repetitive and narrowly scoped subtasks in agentic workloads may be better served by specialized SLMs than by monolithic LLMs. However, the limited capacity and context windows of SLMs ca

Fraglingo: Molecular Design via Attachment-Aware Autoregressive Fragment Generation

arXiv:2609.13519v1 Announce Type: new Abstract: Molecular design is most effective when generation mirrors the edits chemists actually make: extending a scaffold, replacing a substituent, or decorating a scaffold at a specified attachment site while optimizing molecular properties. Fragment-based molecular design naturally supports this workflow, yet existing approaches often separate fragment selection from attachment prediction, first choosing a fragment from a fixed vocabulary and then predicting how it should be connected. This decoupling restricts generation to a closed fragment vocabula

Grounded Adjudication of Variations across Extracted TimeLines (GAVEL): Comparing Clinical Timelines Against Their Case Reports

arXiv:2609.13475v1 Announce Type: new Abstract: Existing pipelines for clinical timeline extraction from case reports are evaluated using an expert reference and are limited by imperfect reference annotations and imprecise event alignment. We developed GAVEL, an LLM judge protocol that compares two timelines with the case report and returns a discrepancy type, verdict, and report passage for each difference. We evaluated the event matcher, reviewed 2,738 findings from GPT5.6sol and DeepSeek V3.2, ranked six LLM extractors and two human annotators, and tested GAVEL guided merging. True match r

AutoTailor: Automatic, User-Aligned Capability Selection and Adaptation for Web Agents

arXiv:2609.13548v1 Announce Type: new Abstract: Web agents can utilize reusable tools to reduce the cost and latency of low-level browser interaction, but automatically discovered tool collections can be large, redundant, and poorly aligned with user demand. We present AutoTailor, a meta-agentic framework for constructing and maintaining a compact set of trajectory-derived Model Context Protocol (MCP) APIs. Offline, AutoTailor converts web trajectories into parameterized browser-automation programs, applies a Quality Filter to remove APIs with unsuitable granularity and redundant functionalit

Asclepius: An Adaptive Harness for Long-Horizon Clinical Agents

arXiv:2609.13543v1 Announce Type: new Abstract: LLM agents are predominantly benchmarked on short, single-task trajectories, yet real deployments run for hours under contention, surfacing a different class of failures. We use the Clinical Environment Simulator (CES), in which an agent manages an entire emergency-department shift under continuous time and resource pressure, as a testbed: long-horizon execution failures manifest measurably in a single rollout under structured, multi-dimensional grading. On CES, current agents reach the correct diagnosis in most cases yet fail to deliver complet

Toward a Decision-Assurance Layer for AI-Assisted Flight Planning in Air Traffic Management

arXiv:2609.13552v1 Announce Type: new Abstract: Generative AI is increasingly being used informally in Air Traffic Management (ATM) for tasks such as flight plan generation, trajectory interpretation, and constraint checking. Although these tools can reduce workload and accelerate planning, their non-deterministic outputs create safety and operational risks in human-in-the-loop settings. This paper proposes the AI Trust and Assurance Layer (ATAL), a model-agnostic decision assurance architecture that evaluates whether AI-generated flight-planning outputs are sufficiently reliable for operatio

Carbon-Aware Routing for Function Calling in Edge-Cloud LLM Systems

arXiv:2609.13559v1 Announce Type: new Abstract: Large Language Models (LLMs) with function-calling capabilities are becoming critical for modern agentic AI systems. Nevertheless, current deployments typically route inferences to powerful cloud-based models, incurring significant energy use and carbon emissions. We address this sustainability challenge with a carbon-aware routing framework that distributes function-calling queries across a three-tier edge-cloud architecture, combining edge and cloud LLMs on heterogeneous hardware. At its core, a lightweight k-NN predictor operating in a unifie

MathematicsprovisionalTier 1 Useful+0.00

ZGCM-1: A Fully Open and Extremely Efficient Foundation Model for Math and Agentic Search

arXiv:2609.13356v1 Announce Type: new Abstract: In this work, we present ZGCM-1, a fully open 7B dense foundation model trained from scratch with extreme data, system, and algorithmic efficiency. ZGCM-1 is founded on a core premise: compact models cannot passively memorize the open web, but can overcome parametric capacity limits by coupling deliberate internal thinking with active external tool use. To support this paradigm across a 256K context, we develop an end-to-end, high-efficiency open training recipe: Architecture & System Co-design: interleaved gated sliding-window and full attentio

TimeThink: Eliciting Compositional Reasoning in Timeseries Large Language Models

arXiv:2609.13457v1 Announce Type: new Abstract: Timeseries multimodal large language models (TS-MLLMs) have recently begun leveraging the reasoning capabilities of large language models (LLMs) for question-answering tasks. However, these models often fail to capture dynamic temporal patterns, providing only implicit reasoning that lacks the underlying explanations critical for high-stakes applications like healthcare. While reinforcement learning (RL)-based timeseries language models aim to address this, they often fall short because they are trained on narrow, in-distribution data and strugg

Accurate and scalable decontamination of imaging-based spatial transcriptomics via optimal transport

Imaging-based spatial transcriptomics enables molecule-resolved profiling of gene expression and tissue organization in situ. However, segmentation errors, transcript spillover and three-dimensional cell overlap can introduce misassigned transcripts into cell-level expression profiles, compromising biological interpretation and obscuring genuine signals. Existing methods either remove suspect expression at the cost of signal loss or lack a biologically grounded criterion for transcript assignment. Here we present CellDot, an optimal-transport framework that determines the fate of each transcri

A Hybrid Agentic AI Framework for Intelligent Supply Chain Analytics

arXiv:2609.13561v1 Announce Type: new Abstract: Efficient utilization of supply chain analytics for decision making remains a significant challenge for planners, as critical tasks such as database querying, key performance indicator (KPI) analysis, demand forecasting, and performance diagnosis require heterogeneous expertise spanning data engineering, operations research, and domain knowledge. In this work, we propose an agentic system for supply chain analytics that bridges the gap between business decision-making and technical expertise, where a coordinator agent interprets user intent and

Expressing glycan motifs as a containment order increases sensitivity and uncovers motif redistributions

Glycan motifs are the standard readout of comparative glycomics, and pipelines treat them as independent features. Here, we show they are not. Motifs are partially ordered by substructure containment, so a parent motif's abundance dominates each of its children. We make that order explicit as a directed acyclic graph, which stratifies the multiple-testing family correctly and, together with an empirical-Bayes variance prior taken from each motif's containment neighborhood, raises estimated true positives across 45 glycomics datasets from 300 to 442 (+47%, p = 0.0001) after controlling for perm

Insight into the mechanism of molecular interactions of the NDM metallo-beta-lactamase and cyclic boronate inhibition by molecular docking and molecular dynamics simulations

The spread and evolution of antimicrobial resistance (AMR) pose significant threats to public health, food safety, and sustainable development. Metallo bea lactamases (MBLs), a class of carbapenemases that hydrolyze carbapenems, often considered last resort antibiotics, are of particular concern and underscore the urgent need to develop effective inhibitors. Recent advancements in bicyclic boronate inhibitors represent significant progress toward this goal. Although the mechanistic basis of NDM1 inhibition by bicyclic boronate inhibitors remains incompletely elucidated, this study systematical

ProteoformTracker: an interactive tool for planning proteoform detectability in top-down and middle-down proteomics

Characterizing proteome complexity in disease contexts is essential for understanding molecular mechanisms and advancing therapeutic development. Mass spectrometry (MS)-based top-down and middle-down proteomics (TDP/MDP) can resolve intact proteoforms - protein molecules carrying a unique combination of isoform sequence and post-translational modifications (PTMs); however, their technical complexity and modest throughput present challenges for experimental planning and limit their broader application. Here, we present ProteoformTracker, an online web tool that prospectively models MS signal an

Harm15 Sep 2026

ReverseScreen.ai: Pharmacophore-Guided Reverse Screening Across the Growing Co-Complex Proteome

Biopharmaceutical companies routinely use forward screening to identify potential ligands for targets of biological consequence. Increasingly, they are also using reverse screening to proactively identify downstream off-target liabilities and new repurposing opportunities. Exhaustive reverse docking of one or a multitude of molecules across every characterized site is one approach, but the computational cost is often prohibitive. One molecule against 28,579 receptor sites takes 40.5 hours on twenty CPU cores. We present a method that puts a retrieval step in front of the docking. Every co-crys

A foundation model learns the sequence and functional grammar of fully human heavy-chain-only antibodies

Antibody language models learn from large natural repertoires, but whether generic representations capture the constraints of specialized antibody formats remains unclear. We first characterized fully human heavy-chain-only antibodies (HCAbs) independently of HCAb-trained models. Source-aware comparisons with conventional human VH domains revealed a reproducible distributional shift localized predominantly to CDR1/2, CDR3 architecture and, where supported, a restricted framework region rather than widespread framework remodeling. These model-independent differences motivated repertoire-specifi

Comparative Genomics of COBRA-like Genes in Theobroma and Herrania Reveals Structural Conservation and Lineage-Specific Variation

COBRA-like (COBL) genes encode plant-specific proteins associated with cell-wall organization, cellulose deposition, and developmental processes. Despite their functional relevance in model plants and crops, this gene family remains poorly characterized in Amazonian Malvaceae species. Here, we performed a comparative genomic analysis of COBL genes in two Theobroma grandiflorum clones, two T. cacao cultivars, and Herrania umbratica. Using homology searches, conserved domain validation, phylogenetic reconstruction, gene structure analysis, chromosomal mapping, transcript abundance profiling, mol

A novel unbiased Linkage Disequilibrium estimator through Random Probing

The linkage disequilibrium (LD) or correlation of alleles at different loci is a fundamental statistic in population genetics. This work presents a new estimator for the standardised LD measure r^2 between a pair of loci. The Random Probe (RP) estimator involves generating a set of random dummy loci, before calculating the r^2 between each synthetic locus and the two focal loci using one of the existing estimators (even it is known to be biased). The correlation between the two vectors of r^2 is the LD estimate. Computer simulations show promising results, with the RP estimator at least as unb

Cell-level random splits leak group-owned answers in single-cell benchmarks

Machine learning models in single-cell biology increasingly forecast differentiation, reprogramming and therapeutic response from early transcriptomic profiles. Testing whether a model has learned real biology requires held-out cells. Single-cell data, however, are grouped: cells from the same clone, patient or batch share the same label. A random split therefore places relatives of each test cell, carrying its label, in the training set, and a model can score well by memorizing a relative instead of learning a transferable rule. Grouped validation removes this leakage but leaves far fewer ind

Adversarial random forests for omics synthesis

Data availability is critical for understanding complex disease pathways and developing robust predictive models. Although high-throughput omics technologies have improved insight into disease mechanisms, data acquisition from inaccessible tissues such as the central nervous system remains a major limitation, causing small sample sizes and complicating early prediction of neurodegenerative disorders such as Alzheimer's and Parkinson's diseases. Generative modeling has emerged as a powerful approach for synthesizing data to support downstream clustering and prediction with small sample size, bu

OtherprovisionalTier 1 Useful+0.03

Token Efficient Task Execution via Application Behavior Modeling for Web Agents

arXiv:2609.13491v1 Announce Type: new Abstract: The strong performance of AI Agents across an impressive variety of tasks is driving an unprecedented investment in agentic infrastructures, however the cost of processing tokens is fast increasing. Web agents automate the execution of web-application tasks described in natural language, by analyzing the web-application's user interface (UI) and interacting with it. This work introduces OdoBot, a novel web-agent architecture that completes tasks at a fraction of the cost when compared to conventional web agents. This is achieved by leveraging a

POME: Graph-based embeddings for partially observed mixed-type data

Partially observed mixed-type (POM) data, as often encountered in clinical, epidemiological, and phenotypic datasets, are very common in biomedical research. Yet, advanced data analysis and machine learning based on POM data is complicated by their heterogeneous nature and often substantial fractions of missing values. While one promising way to overcome these issues is to compute vector-valued embeddings for POM datasets which can then be used for downstream analyses, existing embedding methods are mostly not designed for POM data. To address this gap, we developed POME (partially observed mi

Accounting for pseudo-replication of Linkage Disequilibrium for contemporary Ne estimation

The Linkage Disequilibrium (LD) of unlinked loci can be used to estimate contemporary effective population size (Ne) of one to a few generations ago. In genomic datasets loci on different chromosomes are considered unlinked, but there are many more pairs of unlinked loci than there are independent pairs of chromosomes, resulting to confidence intervals (C.I.) being too narrow if the non-independence is not taken into account. Simulations were run to investigate the correlation structure among LD of unlinked loci, which can be expressed by the LD of loci along the same chromosomes, based on a d

Vibe Patenting: Evaluating LLM Judges for Professional Patent-Drafting Agents

arXiv:2609.13422v1 Announce Type: new Abstract: LLM judges are increasingly used to evaluate and improve AI-generated outputs, yet their reliability for complex professional work remains unclear. We study this problem through Vibe Patenting, an end-to-end patent-drafting testbed for AI-agent evaluation. A separately-invoked LLM judge evaluates generated patent drafts and provides structured feedback for iterative revision. Across multiple inventions and drafting-agent configurations, judge-guided revision consistently improves judge-assessed quality, while unguided revision tends to saturate.

From Legal Text to AI-specific Risk Sources: A Systematic Analysis of the EU AI Act's High-Risk Requirements

arXiv:2609.13535v1 Announce Type: new Abstract: The EU AI Act introduces mandatory requirements for high-risk AI systems with the explicit goal of ensuring the development and operation of trustworthy AI. At the same time, AI risk management practices rely on structured risk taxonomies to systematically identify and treat AI-specific risk sources. As both the AI Act and established risk taxonomies aim to address AI-induced risks, a natural question is whether they align in the risk sources they cover. However, no clear mapping exists between the risks implicitly addressed by the Act's high-ri

‘Stomach bugs’ send protective immune cells rushing from gut to brain

Why AI agents invent their own language if you let them chat

“Weird terms” are a natural—but disturbing—byproduct of agent swarms, researchers find

Climate & EnvironmentprovisionalTier 1 Useful+0.01

Toward Self-Adaptive Physical AI: Can LLM Agents Manage Long-Horizon Physical Tasks?

arXiv:2609.13436v1 Announce Type: new Abstract: Large Language Model (LLM) agents offer a promising path toward autonomously managing long-term physical tasks without human intervention. However, physical tasks require agents to continuously observe the environment, make consequential actions, and remain effective as the environment changes. Existing approaches either require substantial data and retraining, or primarily focus on agents operating in the virtual world. In this work, we explore the feasibility of building a self-adaptive physical AI agent that manages long-term physical tasks i

Governing at Machine Speed: An Adaptive Intelligence Architecture for Real-Time AI Policy Enforcement

arXiv:2609.13466v1 Announce Type: new Abstract: Enterprise AI adoption has reached 78% of organizations globally, yet the infrastructure to govern that adoption has not kept pace. This paper identifies and characterizes the attestation deficit, a structural condition in which organizations maintain governance policies but cannot produce auditable, tamper-evident evidence of enforcement within regulatory timelines. Drawing on empirical data from the Stanford 2026 AI Index Report (362 documented incidents), the IBM/Ponemon 2026 Cost of a Data Breach study (USD 4.99M average cost, 92% lacking ac

Climate & EnvironmentprovisionalTier 1 Useful+0.00

Make AI traceable before it shapes global climate assessments

Medicine & HealthprovisionalTier 1 Useful+0.01

STAT+: New documents reveal problems with Medicare’s prior-authorization pilot

In this edition of STAT Health Tech: The rushed launch of Medicare's AI prior-authorization pilot, policy hints from FDA, ONC officials, and more.

Interpretable Machine Learning Reveals Complementary Age-Related Signatures in the Oral and Gut Microbiome

Whether combining microbiome data from multiple body sites improves prediction, and whether different sites carry complementary or redundant information, are distinct questions that most studies conflate into a single accuracy metric. This work makes two contributions, one methodological and one biological, using paired stool and oral cavity microbiome samples from 44 subjects across two age groups, healthy adults and newborns (Ferretti et al., 2018). Methodologically, we show that a subject-matched fusion design combined with SHAP-based (SHapley Additive exPlanations) site attribution can det

MathematicsprovisionalTier 1 Useful+0.00

Ten numbers from the Laplace-Beltrami spectrum facilitate training-free classification of protein structures based on surface shape

The comparison of protein structures is necessary to determine molecular functions and evolutionary relationships and can also facilitate drug design and the prediction of separation options. The number of available protein structures is rapidly increasing, driven by experimental determination but also prediction methods such as AlphaFold. In this context, traditional structure comparison based on atomic superposition becomes computationally inefficient. To enable scalable analysis and surface shape comparison, alternative approaches represent protein structures using compact, fixed-length vec

STAT+: Medicare’s AI prior authorization pilot was rushed and full of problems, new documents reveal

The rushed launch of a Medicare pilot to use AI for prior authorization led to delayed care in some instances, new documents reveal.

DiffGSP: reversing mRNA diffusion to unlock high-fidelity spatial transcriptomics

Spatial transcriptomics enables gene expression profiling within intact tissues while preserving spatial context, providing unprecedented insights into cellular organization and function. However, mRNA diffusion during tissue processing can cause transcripts originating from adjacent cells to be captured at a given spot. This spatial misalignment challenges a fundamental premise of spatial transcriptomics that measured gene expression faithfully corresponds to its spatial origin, thereby compromising spatial fidelity and potentially biasing biological interpretation. Here, we present DiffGSP,

HeartVar: An LLM-Assisted Tool for Clinical Classification of Variants in Cardiovascular Disease Cohorts

Manual clinical DNA variant classification is the bottleneck of every clinical and research rare disease workflow. The process typically requires a curator to assemble evidence from numerous databases, weigh 28 criteria, reconcile competing evidence, and produce a defensible case for the final classification. Additionally, the framework used to assess variants is not static and successive addenda have revised individual criteria. The most complete and current evidence aggregators available are commercial platforms, which can limit researcher access. We present HeartVar, an open-source web tool

Planning or Learning: Reliability and Cost in Multi-Asset Maintenance

arXiv:2609.13566v1 Announce Type: new Abstract: Industrial maintenance systems involve multiple interacting assets and shared resources, making it challenging to balance reliability and operational cost using a single decision framework. While recent work has focused on reinforcement learning (RL) for maintenance scheduling, direct comparisons with planning approaches under identical settings remain limited. In this work, we empirically compare planning and RL for multi-asset bearing maintenance using run-to-failure data. We examine how these methods behave when balancing preventive maintenan

ImmuneLens: linking transcriptional states and TCR clonotypes through disentangled multimodal learning

Single-cell multi-omics technologies simultaneously capture the transcriptome and TCR sequence of T cells, providing an opportunity to study the relationship between transcriptional states and clonal architectures. However, jointly modeling the relationships between transcriptional states and TCR sequences while preserving modality-specific information remains challenging. Here, we present ImmuneLens, an interpretable multimodal representation learning framework designed for paired single-cell transcriptome and TCR sequence data. ImmuneLens supports the construction of a transferable multi-coh

How Fyxer built an AI executive assistant people trust

Fyxer uses OpenAI models, fine-tuning, memory, and real user feedback to organize inboxes and draft emails in each user’s voice.

LucaCell: a sequence-centric foundation model for cross-species single-cell analysis

Single-cell foundation models have transformed transcriptomic analysis, yet most rely on fixed gene identifiers that limit transfer across species and data types. Here we present LucaCell, a sequence-centric foundation model that represents genes through pre-trained mRNA sequence embeddings rather than static gene annotations. Gene expression is discretized into bins and modeled with a Transformer encoder, enabling sequence-informed cell representation without a fixed gene-ID vocabulary. Pre-training on 85 million human and mouse single cells, LucaCell is evaluated on human, mouse and lemur ge

Drug firms’ secret data supercharge AI protein models

TransBind2: Improving Transcription Factor-DNA Binding Prediction with Multimodal Data and Bidirectional Cross Attention

Accurate genome-wide prediction of transcription factor (TF)-DNA binding remains challenging because many models focus mainly on DNA sequence and overlook chromatin context and TF structure. We previously developed TransBind, a protein-aware model that combines TF and DNA representations through cross-attention. Here, we introduce TransBind2, which improves on TransBind in several ways. It incorporates DNase-seq accessibility and genome mappability tracks as additional input, uses a biomodal protein language model (ProstT5) to capture both TF sequence and structure, and applies bidirectional c

Daily briefing: How this year’s super El Niño is different

T-GADE: Thermodynamical Generative-AI-Driven Evolution of LLM Artifacts

arXiv:2609.12286v1 Announce Type: new Abstract: Integrating evolutionary computation and large language models (LLMs) requires control of population diversity as well as generative capability. Among LLM outputs, those with explicit structure, such as a description paired with code, are structured artifacts; we use artifact for short. We propose T-GADE, which evolves these artifacts by extending thermodynamical genetic algorithms through LLM-based genetic operators and artifact-level diversity evaluation. A common free-energy objective supports generational and steady-state updates, with Fermi

Profiling proteome-level amino acid substitutions in Alzheimer's disease brain tissue

As the most prevalent neurodegenerative disorder worldwide, Alzheimer's disease (AD) remains incompletely understood at the proteome level. In particular, current studies on amino acid (AA) substitutions have predominantly relied on genomic and transcriptomic profiling. Proteome-scale AA substitutions remain largely uncharacterized. Nevertheless, alterations at the DNA and RNA levels cannot fully recapitulate the spectrum of AA substitutions observed at the proteome level. This critical research gap persists largely due to the inherent analytical challenges posed by large-scale proteomic datas

How effective are contrastive learning-based approaches for activity-cliff prediction?

Activity cliffs, defined as structurally similar molecules with vastly different properties represent a fundamental challenge in modern day drug discovery for property prediction models. While Graph Neural Networks (GNNs) have advanced molecular property prediction, they inherently struggle with this problem due to representation collapse and node over smoothening. In this work, we evaluate the efficacy of incorporating various contrastive learning-based loss functions, including the Supervised Contrastive (SupCon) configurations into several GNN backbones to navigate this structure-activity l

AIM: A Privacy-Aware Interoperable Memory Framework for Multi-Agent Multi-User LLM Systems

arXiv:2609.12320v1 Announce Type: new Abstract: Traditional large language models (LLMs) are scoped to individual user sessions, limiting their knowledge to a single conversation and preventing them from learning user preferences that evolve over time. Existing agentic memory systems address this limitation but generally operate at the individual-user level, restricting the public knowledge that could be shared across users to improve downstream responses. We introduce AIM (Agentic Interoperable Memory), a unified, privacy-aware memory framework that enables multi-agent, multi-user LLM system

Resolving context-specific protein-protein interactomes forbiological discovery and therapeutic target prioritisation

Protein function is shaped by cellular context, yet most protein representations and interaction maps remain context-agnostic. Here we present ProtScape, a multiscale graph-learning framework integrating global protein interactions, cell-type gene expression and protein language models to learn context-specific representations and infer interactomes across more than 200 cell types. ProtScape substantially outperforms existing approaches in interaction reconstruction, increasing the area under the precision-recall curve by 40 percentage points. Its predicted interactions were supported by held-

SafetyprovisionalTier 3 Major Harm−0.31

Black Hat USA 2026 | OpenAI's Deep Dive Into Hugging Face Incident - darkreading.com

Black Hat USA 2026 | OpenAI's Deep Dive Into Hugging Face Incident darkreading.com

Dynamics of bleomycin-induced lung injury at single-nucleus resolution

Single-cell RNA sequencing (scRNA-seq) has emerged as a powerful tool for studying individual cell populations in complex tissues during injury and repair. However, it involves extensive sample manipulation, limiting its application in rare and difficult-to-dissociate cell populations. Single-nucleus RNA sequencing (snRNA-seq) bypasses several of these challenges, but its performance in lung disease models is not well characterized. In a longitudinal murine model of bleomycin-induced lung injury, we performed whole lung snRNA-seq to systematically map 46 cell populations and states with minima

STAT+: How $500 explains Republican health policy, and the GLP-1 Rx wave

The one health care policy Republicans agree on, and a look at how rural providers are approaching AI.

A Scalable Distributed-Memory MPI Implementation of Smith-Waterman with Token-Passing Traceback

As genomic sequencing produces increasingly massive datasets, accurate local sequence alignment via the Smith-Waterman(SW) algorithm remains computationally prohibitive due to its space and quadratic time complexity O(mn). While parallelization addresses a path forward, existing MPI-based solutions present a critical bottleneck during the traceback phase, either omitting it entirely or gathering the entire direction matrix to a single root node, which severely limits scalability. To overcome this, we propose a fully distributed Message Passing Interface (MPI) implementation featuring a token p

OtherprovisionalTier 1 Limited Harm−0.00

STAT+: Pharmalittle: We’re reading about another AstraZeneca trial failure, a Novo name change, and more

AstraZeneca disclosed that a breast cancer pill failed in a pivotal trial testing it as a first-line therapy in advanced tumors

MathematicsprovisionalTier 1 Limited Harm−0.00

Population Intelligence: Smallest-Model Numerical Experiment

We introduce a minimal individual based model for population intelligence that retains three structures absent from standard single agent descriptions: a manifold valued intelligence state, endogenous population turnover, and heritable lineage dependent reproductive fitness. The validated S^1 intelligence diffusion was extended with neutral logistic birth death dynamics. Each individual occupies an angle on the circle S^1; noisy mean field attraction produces collective synchronization; logistic birth death dynamics regulate population size; and reproductive rates can depend on lineage and pos

Benchmarking CUT&RUN analysis using motif enrichment

Background. Cleavage under targets and release using nuclease (CUT&RUN) maps the genome-wide locations of chromatin-associated proteins and provides an improved alternative to chromatin immunoprecipitation sequencing (ChIP-seq) for profiling sequence-specific transcription factor binding sites. Identifying these binding sites plays a critical role in understanding gene regulation, and transcription factors provide a useful setting for benchmarking because their well-defined sequence motifs serve as built-in controls for evaluating performance. Compared with ChIP-seq, CUT&RUN achieves higher re

An interpretable peptide-HLA model emergently learns binding energetics and structure

The range of peptides a human leukocyte antigen (HLA) binds and displays modulates immune response and therefore underpins vaccine design, neoantigen discovery, autoimmunity, transplantation, and hypersensitivity reactions. Modern predictors of peptide-HLA (pMHC) binding and presentation are remarkably accurate, but they are black boxes; their internal computations are opaque and post-hoc explanatory methods lack guarantees of attribution. Here we introduce LAtent Motif INteraction Aggregation (LAMINA), an architecture whose prediction is, by construction, interpretable and attributable. LAMIN

PHACTn enables training-free, context-independent inference of nucleotide variant tolerance across the genome

Accurate prediction of single-nucleotide variant (SNV) tolerability across the entire human genome remains a fundamental challenge in computational genomics, particularly for non-coding regions where the regulatory landscape is vast and poorly understood. Machine learning classifiers suffer from data circularity and demographic bias, while genomic language models demand massive computational resources and offer little biological interpretability. Here, we present PHACTn (Phylogeny-Aware Computing of Tolerance for nucleotide variants), a training-free, parameter-minimal method that infers nucle

Learning Symbolic Constraint Representations from Examples: A Neuro-Symbolic Approach

arXiv:2609.12267v1 Announce Type: new Abstract: Learning user-defined concepts as constraint networks has been extensively studied in the constraint acquisition (CA) literature. However, existing approaches typically rely on intensive interactions with a human oracle, making the learning process costly in terms of time and number of queries. In this paper, we propose a neuro-symbolic framework for automatic CA that significantly reduces user involvement by introducing neural Oracle Transformer models which learn to emulate user responses and to generalize conceptual knowledge. Trained on prev

STAT+: GSK touts lung cancer data from two drugs

Late-stage results for Definium's LSD treatment, AstraZeneca breast cancer drug fails key trial, and more biotech news from The Readout

miRAssist: a context-aware, evidence integration framework for interpretable miRNA-target prioritization

Motivation: MicroRNA-target interaction prediction remains challenging because many existing tools provide prediction scores or ranked candidate lists without making the supporting evidence easy to interpret or relate to a specific biological context. Results: Here, we developed miRAssist, a context-aware evidence-integration framework for interpretable miRNA-target prioritization. miRAssist integrates six evidence families, including sequence complementarity, thermodynamic stability, sequence conservation, target-site accessibility, functional binding, and functional repression. A sequence-de

Affective Agent: On-Device Personalized Intervention Reasoning for Wearable Systems

arXiv:2609.12322v1 Announce Type: new Abstract: Affective computing has advanced wearable state inference, but on-device reasoning about whether, when, and how to intervene remains challenging. We present Affective Agent, a three-layer reference architecture for personalized intervention reasoning under uncertainty on wearable-class hardware. It combines a compact sub-billion-parameter language model with physiological evidence, context, and user history to decide whether, when, and how to intervene, without cloud dependency or per-user retraining. The architecture is organized into three int

Opinion: Restoring public trust in science doesn’t have a simple solution. That’s OK

“Messaging can’t substitute for the relational infrastructure needed for building and sustaining trust,” write Michelle Groman and Caroline Montojo.

WinSyn: An Automated Pipeline for Realistic Enterprise Question-Answering Evaluation

arXiv:2609.12171v1 Announce Type: new Abstract: Enterprise settings provide a challenging environment for question-answering agents, which often rely on Retrieval-Augmented Generation, Deep Research (DR), and related techniques. Much of this challenge comes from the complexity of enterprise data: information is often spread across evolving and potentially conflict- ing emails, chat messages, documents, and other artifacts. Existing benchmarks typically have limited real-world complexity, short-form responses, and unnatural queries, so they often fail to capture the challenges of enterprise se

DU-NO: A Parameter-Efficient Double U-Shaped Neural Operator for Phase-Resolving Wave Modeling

arXiv:2609.12115v1 Announce Type: new Abstract: Phase-resolving wave models such as FUNWAVE-TVD are the accuracy standard for nearshore dynamics, resolving the shoaling, refraction, and breaking of individual waves, but their cost rules them out for the ensembles, uncertainty quantification, and real-time warning that operational forecasting demands. Neural operators promise solver-level accuracy at a fraction of that cost, yet on wave-dominated fields the accurate ones are large: hybrid spectral-convolutional operators such as U-FNO (the strongest baseline in our study after DU-NO) buy their

Soft Symbol Grounding for Prototypical Concepts

arXiv:2609.12247v1 Announce Type: new Abstract: Neuro-symbolic models are usually trained with supervision only on final labels, leaving the intermediate concepts unobserved. Since many concept assignments are consistent with a given label, training can predict labels correctly while recovering the wrong concepts, a failure known as a reasoning shortcut. Prototypical networks reduce shortcuts by anchoring each concept to a few labeled examples, but existing methods still couple perception and reasoning through a hand-crafted, task-specific differentiable loss that must be redesigned for every

Hybrid Physics-AI Framework of Body Center of Mass Dynamics from Wrist-Worn Sensors

arXiv:2609.12304v1 Announce Type: new Abstract: Wrist-worn IMU has been widely used for daily-life health monitoring. Yet, it does not fully represent whole-body dynamics, for which the body center of mass (COM) is considered the physiological reference standard. Therefore, this work proposes a simplified kinematic model (KM), which is designed to map the wrist IMU to the COM acceleration. It is built upon several reductive assumptions that enable the solvability of the dynamic equations based on wrist IMU measurements alone. This work further proposes three types of hybrid AI modeling method

Can LLMs in Draft-Verify-Revise Pipelines Resolve Deictic Ambiguity?

arXiv:2609.12162v1 Announce Type: new Abstract: Draft-verify-revise is a common LLM orchestration pattern for scaling inference-time compute. One LLM drafts, a second critiques the draft and provides feedback, and a third uses that feedback to revise the draft into the final output. As context cascades between stages, LLMs at different stages can resolve a context-dependent expression such as "previous" differently. When that happens, the expression undergoes a deictic shift, a change in what it refers to. This phenomenon was studied with a synthetic dataset of 10 base examples, each rendered

Language Is an Insufficient Substrate for Quantitative Reasoning, and Consequential Domains Need Large Quantitative Models

arXiv:2609.12105v1 Announce Type: new Abstract: The prevailing assumption in applied machine learning is that progress on consequential quantitative decisions such as pricing risk, allocating capital, triaging patients, or containing a network intrusion will follow from progress in large language models (LLMs). A language model is trained on a representation of the world that was produced by human description; description is a lossy encoding of the quantitative record, and the loss is irreversible: no downstream model, at any scale, can recover from a description what the description did not

GLARE: Generative Learning via Adversarial Reward Estimation For Social Dynamics Forecasting

arXiv:2609.12165v1 Announce Type: new Abstract: Meeting continuation requires tracking the agenda, speaker roles, participant intentions, and disagreement across long multi-party discussions. We introduce the Meeting Dynamic Forecasting Benchmark (MDFB), constructed from 2,207 real-world meetings and 24,794 future-facing queries. Given a transcript prefix and an active question, a model generates a plausible multi-turn continuation in one call. We evaluate utility---progress toward the question---and human-likeness---plausible conversational flow and role consistency---without requiring exact

Mined from Scientific Literature: Process Schemas for Atomic Layer Deposition and Etching in Materials Science

arXiv:2609.12139v1 Announce Type: new Abstract: Atomic layer deposition (ALD) and atomic layer etching (ALE) are reported heterogeneously across experimental and simulation literature in materials science, hindering comparison and machine-actionable reuse. We present four domain-expert-reviewed JSON Schemas for ALD and ALE experimental and simulation processes. Curated with schema-miner and grounded in QUDT using schema-miner pro, the schemas structure materials, process conditions, configurations , and measured or predicted results. We compare their scope, structure, and semantic grounding,

Single brain implant lets paralyzed people both talk and gesture via an avatar

By decoding two kinds of signals from one brain area, multitasking device offers step toward more natural communication

ForceFlowAb: physics-aware mixture-of-experts flow matching model for antibody CDRs design

Abstract Motivation: Antibodies are a major class of therapeutic molecules, and their recognition of target antigens is largely mediated by complementarity-determining regions (CDRs), making antigen-conditioned CDR design a central problem in antibody engineering. Recent generative methods have enabled antigen-conditioned co-design of CDR sequences and structures, but their limited capacity to capture local interface heterogeneity and lack explicit energy-based guidance during sampling, which may result in unfavorable antibody-antigen interaction energies. Overcoming these limitations requires

‘Multifunctional’ brain implant translates speech and gestures in real time

Competence-Gated Pooling of Language Models and Priors for Event Forecasting

arXiv:2609.12101v1 Announce Type: new Abstract: In hybrid forecasting, a language model is often one of several available signals. A system may already have a market, crowd, or statistical forecast and must decide whether the model adds useful information or should be ignored. The relevant target is therefore not standalone model accuracy, but relative competence, defined as the model's marginal value beyond the available external forecast. Under Brier loss, we characterize when model disagreement can improve an external forecast and derive the gain from using domain-specific rather than glob

Do Influence-Derived Data Perturbations Enable Machine Unlearning? A Controlled Study of Three Plausible Roles

arXiv:2609.12313v1 Announce Type: new Abstract: We evaluate Deep Perturbation Learning (DPL), which perturbs training images and labels along influence-derived directions, in three roles in which prior work has positioned it for machine unlearning: a direct deletion signal (the strongest claim), a utility-preserving regularizer, and a warm start for adversarial unlearning. Evidence for the weaker roles has been used to support the stronger one, so we test each role separately under a matched protocol with exact-seed retraining baselines. An audit of the public implementation identifies two co

SafetyprovisionalTier 1 Limited Harm−0.00

OpenAI's rogue AI agents were attacking a code registry months before the Hugging Face breach - qz.com

OpenAI's rogue AI agents were attacking a code registry months before the Hugging Face breach qz.com

Heterogeneous graph neural networks with biological prior knowledge for interpretable drug repurposing in triple-negative breast cancer

Drug repurposing offers a cost-effective path to new therapies for triple-negative breast cancer (TNBC), a subtype with limited targeted treatment options. We present PRECISION, a framework integrating transcription factor (TF) regulatory networks, protein-protein interactions, and drug-target edges into a heterogeneous graph neural network (GNN) to identify TFs mediating drug sensitivity and prioritize repurposing candidates. The knowledge graph has 23,498 nodes and approximately 830,000 edges from CollecTRI, OmniPath and the PRISM screen. In a fair cell-line hold-out evaluation, the GNN matc

Reading the Whole Heart: Latent-Attention Masked Autoencoders for Multimodal Cardiac Representation Learning

arXiv:2609.12035v1 Announce Type: new Abstract: Cardiovascular diagnosis rests on integrating complementary modalities, like ECG, echocardiography, chest radiographs, and clinical variables, each capturing distinct but correlated aspects of cardiac physiology. Yet most medical foundation models remain modality-specific, combining modalities only for finetuning or post-training. This discards the cross-modal evidence clinicians naturally integrate and ignores the structure within each modality. We introduce Latent-Attention Masked Autoencoders (LAMAE), a multimodal, structure-aware masked auto

Harness or Model? Isolating the Harness Effect in Agentic Coding with a Contamination-Controlled Private Suite

arXiv:2609.11987v1 Announce Type: new Abstract: An agentic coding system couples a language model to a harness: the tools, prompts and control flow that turn a chat model into an autonomous software engineer. Vendors ship harnesses tuned to their own models, and practitioners assume the vendor-native pairing solves more tasks. We measure that assumption with paired same-model contrasts on a private, contamination-controlled suite of 256 repository and post-cutoff contest tasks. The same 80 tasks ran under claude-agent-sdk and under deepagents on claude-opus-4-8, and under the openai-codex SDK

BiologyprovisionalTier 1 Limited Harm−0.01

Harm13 Sep 2026

ABCP_finder: A Transformer Embedding-Based Prediction of Anti-Breast Cancer Peptides

Breast cancer remains one of the leading causes of cancer-related deaths among women worldwide. Drug resistance, toxicity, and limited target specificity are the major challenges in the development of effective therapeutics. Anti-breast cancer peptides (ABCPs) have emerged as effective drug candidate due to its low toxicity, high selectivity, and ability to target cancer. However, it is very time consuming and expensive to identify novel ABCPs only through experiment methods. To address this challenge, we developed ABCP_finder, the first dedicated computational framework specifically designed

Mechanistic Interpretability of Protein Language Models Reveals Encoded Structural and Functional Properties of Intrinsically Disordered Proteins

Protein language models (PLMs) such as ESM-2 encode protein sequences as embeddings for downstream tasks. PLMs are trained on a masked learning objective that leverages evolutionary constraints. While interpretability studies of ESM-2 have focused on folded proteins, their behavior on intrinsically disordered proteins (IDPs), which constitute a substantial fraction of the human proteome and are implicated in numerous diseases, remains understudied. Because IDPs experience different types of evolutionary constraints on their amino acid sequences, we hypothesized that PLMs would behave different

A Framework for Quantifying DNA Methylation Heterogeneity and Detecting Co-methylated loci from Native Nanopore Sequencing

DNA methylation is an important epigenetic mechanism involved in gene regulation. Most methods focus on analysing DNA methylation averaged from multiple reads, yet these average methylation profiles obscure heterogeneity between individual DNA molecules and coordinated methylation states across loci. Native Oxford Nanopore Technology (ONT) sequencing directly captures long, native DNA molecules together with their base modifications, allowing methylation to be studied at single-molecule resolution. Here, we present a scalable framework for genome-wide methylation analysis using ONT sequencing

Iterative Gene Enrichment Analysis: interpretable networks for human and AI-assisted biological insights

Background: Over-representation analysis (ORA) is widely used to interpret gene lists from high-throughput biological experiments. However, ORA often produces long and fragmented lists of enriched terms that are difficult to translate into coherent, testable biological hypotheses. This challenge is amplified by the rapid growth of gene set collections and the need to integrate results across multiple gene-set collections. Large language models (LLMs) can assist in summarizing enrichment outputs, but their effectiveness is limited by the lack of reproducibility and statistically supported repre

Matched full-UDG and non-UDG ancient DNA libraries reveal trade-offs in post-mortem damage correction for imputation and kinship inference

Ancient DNA studies increasingly combine full uracil-DNA glycosylase-treated, full-UDG, and non-UDG libraries, but which computational damage correction to apply before imputation and kinship analysis remains unsettled. We compared terminal trimming, base-quality rescaling and known-SNP masking in matched full-UDG and non-UDG libraries from the same extracts of two medieval Mongolian individuals. All methods reduced or reversed the difference in mean alternative-allele fraction between damage-prone and transversion SNPs but retained different proportions of covered sites. In non-UDG libraries,

AlphaPeptTools: scverse-native analysis of mass spectrometry-based proteomics

Mass spectrometry (MS)-based proteomics now routinely profiles proteomes at scale, but extracting biological insight and integrating complementary modalities remains challenging. Here, we present AlphaPeptTools, a Python package for the analysis of MS-proteomics data built on the AnnData data structure of the scverse. AlphaPeptTools implements scalable and modular strategies for MS-data ingestion, quality control, preprocessing, and statistical analysis while seamlessly integrating with the scverse ecosystem, unlocking multi-level, spatial, and multimodal analyses.

Harm13 Sep 2026

Machine Learning-Guided Classification of Druggable Pockets and Phylogenetic Druggability Transfer Across the Human Kinome

Protein kinases are among the most intensively pursued therapeutic targets in oncology and beyond, yet selectivity remains largely unsolved; around 536 Manning kinase domains share a conserved ATP-binding pocket, making it difficult to target one without hitting others. Allosteric binding modes which exploit conformational states unique to individual kinases or narrow kinase subfamilies offer a principled route to selectivity. Yet the absence of a kinome-wide structural landscape of these pockets limits their translational potential. Here we construct a machine learning-guided structural atlas

Medicine & HealthprovisionalTier 1 Useful+0.01

A structural antibody benchmark for leakage-aware deep-learning evaluation

Deep-learning methods for antibody structure prediction, antibody-antigen interaction modelling and design are advancing rapidly. However, comparisons across studies remain difficult because training and test sets are often constructed independently, and a temporal cutoff alone does not prevent train-test leakage. We present SABLE (Structural Antibody Benchmark for deep-Learning Evaluation), a versioned structural antibody resource that couples a fixed training collection with a leakage-controlled held-out test set for reproducible machine-learning development and evaluation. SABLE combines 16

HARIBOSS++: An Integrated Platform for RNA-Targeted Drug Design

Targeting RNA with small molecules is increasingly viewed as a pivotal strategy for future therapeutics. A major limitation in studying RNA-small molecule (RNA-SM) interactions is the lack of available experimental structures. HARIBOSS established a milestone as a curated database of RNA-SM complexes derived from the Protein Data Bank. Building on its community engagement, we present HARIBOSS++, an updated and next-generation platform designed to facilitate the study of future RNA-targeted drugs. Major updates include an extension of the pocketome to encompass binding sites where small molecul

Benefit11 Sep 2026

Briefing Chat: The Bunsen burner myth that turns out to be just hot air

Benefit11 Sep 2026

Daily briefing: Skeleton teeth show evidence of the earliest known habitual drug use

MathematicsprovisionalTier 1 Useful+0.02

Benefit10 Sep 2026

Here is how to understand OpenAI’s major mathematical breakthrough - New Scientist

Here is how to understand OpenAI’s major mathematical breakthrough New Scientist

Benefit10 Sep 2026

Daily briefing: Text-message tutoring keeps kids learning when schools close

Benefit10 Sep 2026

AI researchers reckon with the $1.5 million ‘academia tax’

Benefit9 Sep 2026

How AI is changing PhD programmes: join our free webinar

Benefit9 Sep 2026

Within-family effect of ancestry on complex traits in a Mexican population

Medicine & HealthprovisionalTier 3 Major+0.25

Benefit24 Aug 2026

Protege DataLab associates faster ambient AI adoption with lower 2026 in-system mortality

A Protege DataLab / a16z chart, recirculated by Hedgeye on 20 September 2026, plots in-system mortality per 100k patients by ambient AI adoption (share of notes referencing AI tools). High- and low-adoption systems tracked together from 2023 through 2025, then the 2026* points diverge to about 1,010 versus 1,620. Error bars are wide, 2026 is incomplete, no named model is identified, and the authors frame related network figures as descriptive rather than causal.

Harm27 Jul 2026

Authors have mixed feelings about the $1.5B Anthropic copyright infringement ruling - NPR

Authors have mixed feelings about the $1.5B Anthropic copyright infringement ruling NPR

Benefit3 Jul 2026

Google DeepMind and A24 announce first-of-its-kind research partnership