What happened
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 correctness issues: image directions are computed on augmented, normalized tensors but applied to raw images, and the label perturbation falls below float32 resolution, leaving labels unchanged. After correcting the image-perturbation pipeline, DPL fails the direct-deletion criterion on CIFAR-10/ResNet-18 in all three paired seeds. Its utility effects are inconsistent in sign across seeds, and once direction-computation time is counted it underperforms simple warm-start baselines. A one-seed Tiny ImageNet check likewise does not favor DPL as a regularizer or warm start; preprocessing inconsistencies in the released code make the direct comparison there inconclusive. These results cover random instance deletion only and do not rule out influence-based methods in other deletion regimes. We release a role-matched evaluation protocol and an audit checklist for perturbation-based deletion claims. Update (2024-10-09): New Support Through The Audacious Project October 9, 2024 Funding for Canary will enable research and implementation at scale Read more. New Support Through The Audacious Project October 9, 2024 Funding for Canary will enable research and implementation at scale Read more Update (2026-09-16): Cross-Anatomy Transfer Versus Sparse Interpolation in Digital-Twin-Oriented Aortic Fluid-Structure Interaction Surrogates. arXiv:2609.16322v1 Announce Type: new Abstract: Surrogate credibility for fluid-structure interac- tion (FSI) requires distinguishing transfer across independent anatomies from interpolation within an already sampled surface. Four de-identified human aortic models from the Vascular Model Repository were reconstructed into separate lumen and nominal 1.5-mm wall domains and analyzed under matched first-cycle two-way FSI. A geometry-only LightGBM prior, selected by leave-one-anatomy-out development on three anatomies, was zero-shot evaluated on a fourth, then probed with a post-zero- shot sparse field-completion case study over six targets. Zero-shot transfer was poor across all targets. At a five-percent anchor level (203 anchors, 3,852 evaluation nodes), prior-plus-adaptation reached an oscillatory shear index (OSI) R2 of 0.603. However, same-anchor controls tuned only on the three development anatomies were stronger for several outcomes: inverse-distance weighting reached R2 = 0.829 (OSI), 0.617 (peak von Mises stress), 0.676 (mean stres Update (2026-06-18): Reinforcement learning towards broadly and persistently beneficial models. Training targeting beneficial behavior in realistic scenarios produces broad improvements in alignment that generalize across domains and persist under adversarial pressure. Update (2026-09-17): Making AI-Assisted Claims Independently Challengeable: Publication Authority and a Protocol for Falsifiable Publication Records. arXiv:2609.17631v1 Announce Type: new Abstract: AI-assisted claims can appear authoritative when evidence, analysis, human authorization, presentation, and correction history refer to different states. Provenance, attestation, and transparency expose history but alone do not specify the publication transition examined here. We develop Publication Authority as an exact-state, non-transferable, single-use publication capability and instantiate it in PAC-2026 (Publication-Accountability Calculus), a machine-readable AIJIM Protocol candidate. We evaluate its fourth bounded semantic freeze (SF-4), a fixed-profile specification designed for replaceable bindings. Six obligations govern evidence, runs and artifacts, measurement disclosure, authorization, surface correspondence, and lifecycle continuity. Each yields a target-bound witness, localized counterexample, or localized unverifiability; none can compensate for another. Only a fresh, complete all-pass record derives the permit consumed by one atomic publication transition. We use ide Update (2026-09-17): Imitation Learning for Autonomous Driving in CARLA. arXiv:2609.17757v1 Announce Type: new Abstract: Behavioral cloning trains a policy offline on expert demonstrations, but deployment is closed loop: each action affects the observations the policy receives next. We study how much closed-loop driving competence a compact multimodal policy can acquire from offline demonstrations in the CARLA simulator. The policy uses five-frame histories of RGB images, LiDAR, vehicle telemetry, and lane waypoints to predict throttle, brake, and steering at 20 Hz. Demonstrations were collected in three stages, ending with a systematic route-generation procedure that enumerates spawn points and feasible maneuvers and verifies completed autopilot routes. The released 1.36 million parameter policy was trained on 236,882 windows, representing about 3.3 hours of driving from 448 captures. The resulting policy drives autonomously for hours on training and held-out routes. In our runs, it did so without collisions and also transferred qualitatively to an unseen CARLA town with different road geometry. We also observed recovery from large trajectory deviations, although we do not clai Update (2026-09-21): LoRA Enhanced Contrastive Learning with SAS Vision Transformers. arXiv:2609.21061v1 Announce Type: new Abstract: Automatic target recognition (ATR) with synthetic aperture sonar (SAS) supports advanced naval capabilities, but deep learning is constrained by scarce target imagery, background clutter, and human-in-the-loop assessment. We adapt DINOv3 Vision Transformer (ViT) models to underwater SAS ATR using a three-stage parameter-efficient framework. Stage 1 uses Low-Rank Adaptation (LoRA) while freezing the ViT backbone, bridging the gap between natural-image pretraining and underwater acoustic propagation. Stage 2 uses hard-negative mining to strengthen the decision boundary against acoustic mimics, including rocks and sediment formations resembling man-made targets. Stage 3 uses Supervised Contrastive Learning (SupCon) to separate target and clutter representations. We evaluate at-sea SAS data using a mission-level geographic split, compare all arms at 85 percent test recall, and repeat each comparison over three random seeds. LoRA accounts for the primary effect, increasing area under the precision-recall curve (AUPRC) from 0.300 to 0.679 +/- 0.027 usin Update (2026-09-22): PAANI : On Device Visual Evidence Fusion and Explainable Guidance for River Robot Simulation. arXiv:2609.22353v1 Announce Type: new Abstract: Mobile river monitoring robots must interpret obstacles and water boundaries that geographic waypoints alone cannot describe. On resource constrained platforms, converting imperfect visual predictions into timely and inspectable guidance is a distinct challenge. An object label or steering command does not explain which evidence supports a decision or when that evidence is unreliable. We present PAANI, an on-device perception to guidance architecture that combines a project trained YOLO11n detector and a custom MobileNetV3 Small semantic segmenter with timestamp aligned evidence fusion on Arduino UNO Q. Bounded tracking supplies object persistence, while an explicit corridor policy combines surface labels, accepted detections, urgency and mask uncertainty. Each final advisory exposes its contributing evidence and policy reasons. ROS 2 interfaces connect the local AI pipeline to a separate Gazebo vessel, localization and control testbed. Training uses 10,000 WaterScenes images for four-class detection and 1,127 MaSTr132 Update (2026-09-23): From Offline Proxies to Online Decisions: A Layered Engagement Evaluation Framework for Conversational AI. arXiv:2609.25408v1 Announce Type: new Abstract: Online A/B experiments are the decision standard for user engagement, but traffic and readout time limit how many conversational-AI changes can be tested. We ask whether an offline signal designed to be computable without treatment-arm user exposure agrees with the outcomes of those experiments. We contribute a reusable construction and diagnosis checklist that treats an offline proxy as a chain of three alignments: behavioral label to product outcome, learned classifier to candidate-assistant behavior, and aggregated offline signal to experiment effect. A companion evaluation protocol audits the whole composite by interval-aware decision agreement, which compares offline and online confidence intervals instead of point estimates, and by within-experiment ranking. The instantiation we evaluate comprises a fixed evaluation suite on which candidate behavior is scored, an engagement classifier trained to predict session/prompt level engagements, and a calibration layer mapping sample-level score differences t Update (2026-09-23): Open-world fungal ITS embeddings improve higher-rank placement and cross-view retrieval, but percent identity is stronger on a matched ITS2 novelty benchmark . 1. Background. Fungal internal transcribed spacer (ITS) analyses usually infer taxonomy and novelty from similarity to homologous reference sequences. Percent identity is a strong signal in that setting, but it cannot directly compare non-homologous barcode views and does not explicitly optimize the higher-rank placement of genera absent from the reference collection. We asked whether a compact sequence encoder could learn that open-world structure and whether maximum embedding cosine similarity could improve novelty detection. 2. Design. From the UNITE dynamic release of 19 February 2025 we built ITS-core, ITS1 and ITS2 views and a genus-separated development, calibration and test design. A convolutional encoder mapped sequences to 256-dimensional L2-normalized embeddings. A controlled ladder compared genus-proxy supervision (M0), cross-view invariance (M1), hierarchical taxonomic geometry (M2), leave-one-genus family episodes (M3), and the combined objective (M4). M4 was frozen before CAL_KNOWN a Update (2026-09-24): CycDelta: Multimodal Residue-Graph Learning of Permeability Changes in Cyclic Peptides . Membrane permeability remains a major challenge in cyclic-peptide drug discovery, where optimization often depends on estimating changes relative to measured reference compounds. Existing predictors mainly model absolute permeability, while heterogeneity across literature sources complicates pooled training and comparative evaluation. We present CycDelta, a multimodal residue-graph model trained directly on within-source permeability differences. CycDelta combines directed message passing with pretrained Uni-Mol monomer embeddings, physicochemical descriptors and assay information in a shared encoder for reference and query peptides. To evaluate this comparative formulation under controlled conditions, we establish CycDeltaBench, comprising in-distribution, scaffold-based out-of-distribution, external and zero-shot unseen-assay tests. On the out-of-distribution task, CycDelta achieved Pearson 0.69 versus 0.46 for the strongest baseline. It also led all evaluated methods on two external data sets and achieved zero-shot Spearman 0.70 versus 0.32 for the strongest baseli Update (2026-09-24): FedEdgeR: federated and privacy-preserving edgeR for differential gene expression analysis . Motivation: Multi-center RNA-seq studies improve statistical power, but privacy regulations restrict patient-level data sharing. Meta-analysis methods avoid this restriction but lose power, especially under per-site imbalance. Federated learning allows sites to share only summary statistics. Among the three dominant differential expression (DE) frameworks, Flimma federates limma-voom and FedPyDESeq2 federates DESeq2. edgeR, still preferred for small-sample and high-variability designs, lacks a federated counterpart. Its iteratively reweighted least squares (IRLS) and Cox-Reid dispersion estimation make federation harder than limma-voom's single-pass fit. Results: We present FedEdgeR, a federated edgeR implementation protected by secure multi-party computation (SMPC), covering IRLS GLM fitting, three-level dispersion estimation, and the likelihood-ratio test. We evaluate it on four representative RNA-seq datasets: two tumor/normal cohorts, a multi-cohort anti-PD-1 immunotherapy study, and a 6-sample paired stress-test. FedEdgeR matches pooled edgeR under various me