{"data":{"event":{"id":"7cff27ff-597e-4136-a96b-a031619371fc","slug":"token-efficient-task-execution-via-application-behavior-modeling-for-web-aea8cd1f63","title":"Token Efficient Task Execution via Application Behavior Modeling for Web Agents","short_summary":"arXiv:2609.13491v1 Announce Type: new \nAbstract: 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 ","full_description":"arXiv:2609.13491v1 Announce Type: new \nAbstract: 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 behavioral model of the underlying application constructed by analyzing successful task-execution demonstrations. Our experiments with 45 tasks on the Canvas Learning Management System (LMS) demonstrate that OdoBot uses 44% and 80% fewer tokens than two state-of-the-art competitor agents (Agent-E and WebVoyager), while also surpassing WebVoyager in terms of task success rate.\n\nUpdate (2026-08-31): GigaPath-Flash and GigaTIME-Flash: Toward population-scale discovery with efficient pathology foundation models. What if pathology foundation models could do more with less? GigaPath-Flash and GigaTIME-Flash cut computational demands while maintaining strong performance, opening the door to larger studies and broader exploration.\nThe post GigaPath-Flash and GigaTIME-Flash: Toward population-scale discovery with efficient pathology foundation models appeared first on Microsoft Research.\n\nUpdate (2026-08-03): Orchard: An open framework for scalable agentic AI. Orchard is an open-source framework for the research community to train and evaluate AI agents across task types. It reduces complexity while supporting strong performance from smaller models by enabling researchers to reuse the same infrastructure.\nThe post Orchard: An open framework for scalable agentic AI appeared first on Microsoft Research.","ledger_type":"benefit","primary_domain_id":"fc7d617f-c9df-4b9f-b595-4408915ea380","event_status":"provisional","event_date":"2026-09-15T00:00:00.000Z","discovery_date":"2026-09-15T00:00:00.000Z","first_published_date":"2026-09-15T00:00:00.000Z","last_reviewed_date":"2026-08-03T00:00:00.000Z","geographic_scope":"International","affected_population":null,"base_impact_tier":1,"base_score":"1.00","attribution_multiplier":"0.5000","evidence_multiplier":"0.3000","realization_multiplier":"0.4000","durability_multiplier":"0.5000","current_event_score":"0.030000","confidence_level":"low","score_explanation":"Ingest update: raised attribution; independent source raised evidence; raised realization; added a follow-up source. Score remains provisional until review. Follow-up ingest: independent source raised evidence; added a follow-up source.","methodology_version_id":"8745541b-0f8e-4d5a-9b47-5291d3813a8d","original_methodology_version_id":"8745541b-0f8e-4d5a-9b47-5291d3813a8d","published_at":"2026-09-15T04:00:55.905Z","created_at":"2026-09-15T04:00:55.905Z","updated_at":"2026-09-17T03:01:05.630Z","flags":[],"domain_name":"Other","domain_slug":"other","methodology_version":"0.2"},"contributions":[{"id":"c8f5b711-e4ad-4227-b6e0-6c26ac851025","event_id":"7cff27ff-597e-4136-a96b-a031619371fc","model_id":"535014f7-c941-4d39-aa9b-8f9c1d3d07b7","role_description":"Unspecified system mentioned or implied by a news item. Remap to a named release when identified.","attribution_multiplier":"0.5000","credit_share":"1.0000","contribution_score":"0.030000","attribution_rationale":"News ingest does not infer a named model from the publisher alone. Attribution stays unspecified until a release is identified.","attribution_confidence":"medium","first_used_date":"2026-09-15T00:00:00.000Z","model_version_if_known":null,"review_status":"approved","created_at":"2026-09-15T04:00:55.914Z","model_slug":"unspecified-ai-system","model_name":"Unspecified AI system","identity_class":"unknown","is_internal":false,"family_name":"Unspecified","family_slug":"unknown-unspecified","organization_name":"Unknown","organization_slug":"unknown"}],"sources":[{"id":"8c358395-9b8f-4a0e-9f39-3fffe076e968","event_id":"7cff27ff-597e-4136-a96b-a031619371fc","url":"https://arxiv.org/abs/2609.13491","canonical_url":"https://arxiv.org/abs/2609.13491","source_type":"preprint","publisher":"arXiv cs.AI","author":null,"publication_date":"2026-09-15T00:00:00.000Z","retrieved_at":"2026-09-15T04:00:55.921Z","title":"Token Efficient Task Execution via Application Behavior Modeling for Web Agents","excerpt":"arXiv:2609.13491v1 Announce Type: new \nAbstract: 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 a","content_hash":null,"source_reliability_class":"medium","is_primary_source":true,"is_independent":true,"is_peer_reviewed":false,"archived_url":null,"created_at":"2026-09-15T04:00:55.921Z"},{"id":"476bd963-b1d0-4781-8253-8b895f1074dc","event_id":"7cff27ff-597e-4136-a96b-a031619371fc","url":"https://www.microsoft.com/en-us/research/blog/orchard-an-open-framework-for-scalable-agentic-ai/","canonical_url":"https://www.microsoft.com/en-us/research/blog/orchard-an-open-framework-for-scalable-agentic-ai/","source_type":"developer_announcement","publisher":"Microsoft Research","author":null,"publication_date":"2026-08-03T00:00:00.000Z","retrieved_at":"2026-09-17T03:01:05.617Z","title":"Orchard: An open framework for scalable agentic AI","excerpt":"Orchard is an open-source framework for the research community to train and evaluate AI agents across task types. It reduces complexity while supporting strong performance from smaller models by enabling researchers to reuse the same infrastructure.\nThe post Orchard: An open framework for scalable agentic AI appeared first on Microsoft Research.","content_hash":null,"source_reliability_class":"medium","is_primary_source":false,"is_independent":true,"is_peer_reviewed":false,"archived_url":null,"created_at":"2026-09-17T03:01:05.617Z"},{"id":"553c463b-7f64-4c9e-a07f-077e63ed3472","event_id":"7cff27ff-597e-4136-a96b-a031619371fc","url":"https://www.microsoft.com/en-us/research/blog/gigapath-flash-and-gigatime-flash-toward-population-scale-discovery-with-efficient-pathology-foundation-models/","canonical_url":"https://www.microsoft.com/en-us/research/blog/gigapath-flash-and-gigatime-flash-toward-population-scale-discovery-with-efficient-pathology-foundation-models/","source_type":"developer_announcement","publisher":"Microsoft Research","author":null,"publication_date":"2026-08-31T00:00:00.000Z","retrieved_at":"2026-09-17T03:01:05.396Z","title":"GigaPath-Flash and GigaTIME-Flash: Toward population-scale discovery with efficient pathology foundation models","excerpt":"What if pathology foundation models could do more with less? GigaPath-Flash and GigaTIME-Flash cut computational demands while maintaining strong performance, opening the door to larger studies and broader exploration.\nThe post GigaPath-Flash and GigaTIME-Flash: Toward population-scale discovery with efficient pathology foundation models appeared first on Microsoft Research.","content_hash":null,"source_reliability_class":"medium","is_primary_source":false,"is_independent":true,"is_peer_reviewed":false,"archived_url":null,"created_at":"2026-09-17T03:01:05.396Z"}],"claims":[{"id":"a4fa41cc-4020-4a37-ae19-76ea9b66d3d4","event_id":"7cff27ff-597e-4136-a96b-a031619371fc","claim_text":"Token Efficient Task Execution via Application Behavior Modeling for Web Agents","claim_type":"outcome","claim_status":"supported","confidence_score":null,"created_at":"2026-09-15T04:00:55.932Z","updated_at":"2026-09-15T04:00:55.932Z"}],"revisions":[{"id":"58d34578-1f84-4f8f-9625-93b10a687030","event_id":"7cff27ff-597e-4136-a96b-a031619371fc","model_id":null,"previous_score":"0.000000","new_score":"0.001000","previous_factors":{},"new_factors":{"evidence":0.1,"base_score":1,"durability":0.5,"attribution":0.1,"realization":0.2},"change_reason":"Auto-published from news ingest.","trigger_type":"news_ingest","trigger_source_ids":null,"reviewer_id":null,"review_status":"published","created_at":"2026-09-15T04:00:55.938Z"},{"id":"3600534d-b7fc-42ef-b665-783eec5bba36","event_id":"7cff27ff-597e-4136-a96b-a031619371fc","model_id":null,"previous_score":"0.001000","new_score":"0.020000","previous_factors":{"tier":1,"evidence":0.1,"base_score":1,"durability":0.5,"attribution":0.1,"realization":0.2},"new_factors":{"tier":1,"models":["535014f7-c941-4d39-aa9b-8f9c1d3d07b7"],"evidence":0.2,"base_score":1,"durability":0.5,"attribution":0.5,"realization":0.4},"change_reason":"Ingest update: raised attribution; independent source raised evidence; raised realization; added a follow-up source.","trigger_type":"ingest_update","trigger_source_ids":["553c463b-7f64-4c9e-a07f-077e63ed3472"],"reviewer_id":null,"review_status":"published","created_at":"2026-09-17T03:01:05.418Z"},{"id":"a54ef72f-b28f-425a-8712-7fd993dc64d5","event_id":"7cff27ff-597e-4136-a96b-a031619371fc","model_id":null,"previous_score":"0.020000","new_score":"0.030000","previous_factors":{"tier":1,"evidence":0.2,"base_score":1,"durability":0.5,"attribution":0.5,"realization":0.4},"new_factors":{"tier":1,"models":["535014f7-c941-4d39-aa9b-8f9c1d3d07b7"],"evidence":0.3,"base_score":1,"durability":0.5,"attribution":0.5,"realization":0.4},"change_reason":"Ingest update: independent source raised evidence; added a follow-up source.","trigger_type":"ingest_update","trigger_source_ids":["476bd963-b1d0-4781-8253-8b895f1074dc"],"reviewer_id":null,"review_status":"published","created_at":"2026-09-17T03:01:05.636Z"}],"secondary":[]},"methodology_version":"0.1"}