{"paper":{"title":"Learning from Research: Toward Lifelong Agent Harness Evolution","arxiv_id":"2609.40169","primary_cat":"cs.AI","submitted_at":"2026-09-30T17:00:38+00:00","version_compiled":"v1"},"notice":"Every quoted sentence is the authors' text from the arXiv listing abstract. Labels and probabilities are the outputs of classifier-1.13.0 on the question bank abstract-p1-20260923, exactly as returned. No sentence in this record was written by a language model.","schema":"pith.compiled-record/v0","status":"compiled","abstract":{"source":"arXiv listing abstract v1","sentences":[{"i":0,"p":1.0,"id":"s0","own":false,"role":"background","text":"Language agents are expected to solve increasingly complex tasks, creating a growing need for continual improvement.","hedge":{"p":0.7,"label":"asserted"},"own_p":0.05,"main_claim_p":0.02,"quantitative":false,"quantitative_p":0.06,"role_probabilities":{"gap":0.0,"method":0.0,"result":0.0,"background":1.0,"limitation":0.0,"future_work":0.0,"contribution":0.0,"interpretation":0.0}},{"i":1,"p":0.85,"id":"s1","own":false,"role":"background","text":"One promising approach is to evolve the agent harness, the software that governs tool use, memory management, and task execution, while keeping the underlying language model fixed.","hedge":{"p":0.86,"label":"hedged_possibility"},"own_p":0.07,"main_claim_p":0.06,"quantitative":false,"quantitative_p":0.04,"role_probabilities":{"gap":0.0,"method":0.11,"result":0.0,"background":0.85,"limitation":0.0,"future_work":0.0,"contribution":0.04,"interpretation":0.0}},{"i":2,"p":0.99,"id":"s2","own":false,"role":"background","text":"Recent methods automate this process by using a meta coding agent to modify the harness based on execution feedback.","hedge":{"p":1.0,"label":"asserted"},"own_p":0.06,"main_claim_p":0.03,"quantitative":false,"quantitative_p":0.03,"role_probabilities":{"gap":0.0,"method":0.01,"result":0.0,"background":0.99,"limitation":0.0,"future_work":0.0,"contribution":0.0,"interpretation":0.0}},{"i":3,"p":0.85,"id":"s3","own":false,"role":"gap","text":"However, relying on that agent's existing knowledge and observed failures can restrict exploration and make adaptation reactive.","hedge":{"p":0.55,"label":"asserted"},"own_p":0.1,"main_claim_p":0.03,"quantitative":false,"quantitative_p":0.04,"role_probabilities":{"gap":0.85,"method":0.0,"result":0.0,"background":0.11,"limitation":0.04,"future_work":0.0,"contribution":0.0,"interpretation":0.0}},{"i":4,"p":1.0,"id":"s4","own":true,"role":"contribution","text":"Inspired by how human experts learn from the research literature for new solutions, we introduce ScholarEvolve, a framework that automatically draws on state-of-the-art research to guide harness evolution.","hedge":{"p":1.0,"label":"asserted"},"own_p":0.86,"main_claim_p":0.17,"quantitative":false,"quantitative_p":0.03,"role_probabilities":{"gap":0.0,"method":0.0,"result":0.0,"background":0.0,"limitation":0.0,"future_work":0.0,"contribution":1.0,"interpretation":0.0}},{"i":5,"p":0.97,"id":"s5","own":true,"role":"method","text":"ScholarEvolve organizes the harness evolution directions into functional modules and uses topic modeling to identify distinct improvement strategies for each module.","hedge":{"p":1.0,"label":"asserted"},"own_p":0.87,"main_claim_p":0.06,"quantitative":false,"quantitative_p":0.03,"role_probabilities":{"gap":0.0,"method":0.97,"result":0.0,"background":0.0,"limitation":0.0,"future_work":0.0,"contribution":0.03,"interpretation":0.0}},{"i":6,"p":0.98,"id":"s6","own":true,"role":"method","text":"It implements these strategies and evaluates their combinations to improve task performance.","hedge":{"p":0.99,"label":"asserted"},"own_p":0.88,"main_claim_p":0.08,"quantitative":false,"quantitative_p":0.03,"role_probabilities":{"gap":0.0,"method":0.98,"result":0.0,"background":0.0,"limitation":0.0,"future_work":0.0,"contribution":0.02,"interpretation":0.0}},{"i":7,"p":0.59,"id":"s7","own":true,"role":"contribution","text":"Moreover, the framework is designed to incorporate new publications over time, allowing research advances to drive proactive lifelong evolution.","hedge":{"p":1.0,"label":"asserted"},"own_p":0.58,"main_claim_p":0.1,"quantitative":false,"quantitative_p":0.03,"role_probabilities":{"gap":0.0,"method":0.39,"result":0.0,"background":0.0,"limitation":0.0,"future_work":0.02,"contribution":0.59,"interpretation":0.0}},{"i":8,"p":0.99,"id":"s8","own":true,"role":"result","text":"Experiments demonstrate improvements on AppWorld and Tau2-Bench.","hedge":{"p":1.0,"label":"asserted"},"own_p":0.93,"main_claim_p":0.24,"quantitative":false,"quantitative_p":0.04,"role_probabilities":{"gap":0.0,"method":0.0,"result":0.99,"background":0.0,"limitation":0.0,"future_work":0.0,"contribution":0.01,"interpretation":0.0}},{"i":9,"p":1.0,"id":"s9","own":true,"role":"result","text":"ScholarEvolve raises Qwen3.5-27B task goal completion from 49.6% to 63.6% on AppWorld Challenge, and raises GPT-5.4-mini pass@1 from 72.7% to 81.9% on Tau2-Bench Telecom.","hedge":{"p":1.0,"label":"asserted"},"own_p":0.94,"main_claim_p":0.38,"quantitative":true,"quantitative_p":0.99,"role_probabilities":{"gap":0.0,"method":0.0,"result":1.0,"background":0.0,"limitation":0.0,"future_work":0.0,"contribution":0.0,"interpretation":0.0}}]},"compiled":{"main_claim":{"p":0.54,"text":"Experiments demonstrate improvements on AppWorld and Tau2-Bench.","location":"abstract s8","hedge_strength":{"p":1.0,"label":"asserted"}},"provenance":{"route":{"via":"TypeSafe API","model":"classifier-1.13.0","endpoint":"https://api.typesafe.ai/v1/systemone","provider":"TypeSafe"},"packing":"one request per paper","requests":1,"input_tokens":7502,"cost_usd_reported":0.0},"compiled_at":"2026-10-01","bank_version":"abstract-p1-20260923","model_version":"classifier-1.13.0","abstract_states":{"claims_first":0.06,"states_limitation":0.12,"human_participants":0.03,"controlled_comparison":0.56,"randomized_assignment":0.02,"states_quantitative_result":0.99,"compares_with_prior_results":0.85,"mentions_code_or_data_availability":0.04},"evidential_basis":{"p":0.99,"value":"observation or experiment","probabilities":{"argument":0.0,"literature":0.0,"proof_or_derivation":0.0,"analysis_of_existing_data":0.0,"observation_or_experiment":0.99,"simulation_or_computation":0.01}},"contribution_type":{"p":1.0,"value":"method or tool","probabilities":{"review":0.0,"method_or_tool":1.0,"empirical_study":0.0,"simulation_study":0.0,"theoretical_result":0.0,"dataset_or_benchmark":0.0,"position_or_commentary":0.0}},"contribution_statement":{"p":1.0,"text":"Inspired by how human experts learn from the research literature for new solutions, we introduce ScholarEvolve, a framework that automatically draws on state-of-the-art research to guide harness evolution.","location":"abstract s4"}},"coverage":{"tier":"abstract","levels_run":["P abstract"],"not_examined":["full text","figures and tables","references and citing sentences","relations to other papers"],"sources_read":["arXiv listing abstract v1"],"levels_not_run":["S sections","T body sentences","C citations","R cross-paper relations"],"sentences_examined":10,"sentences_in_abstract":10},"measured_accuracy":{"url":"https://pith.science/accuracy","measured_on":"2026-10-02","method":"two independent model families labelled a fresh random sample blind; the figure is how often the published label matches their consensus","sentence_role_agreement":0.92,"main_claim_agreement":0.908,"typed_citation_agreement":0.933,"typed_citation_agreement_95_interval":[0.917,0.945],"review_comments_accurate":0.758,"review_comments_accurate_95_interval":[0.684,0.82],"review_caught_admitted_error_found_or_related":0.448,"review_comments_judged":{"judged_by":["gpt-5.6-luna","grok-4.7"],"judged_against":"Pith's own text copy of each paper: text extracted from the PDF, the first 90,000 characters","measured_on":"2026-10-04","accurate":113,"decided":149,"sampled_reviews":200,"split_by_lane":false,"by_lane":"https://pith.science/accuracy#by-lane"}}}