{"paper":{"title":"SelfEvoSkill: Paired Execution Audits for Skill Revision without External Outcome Supervision","arxiv_id":"2606.14239","primary_cat":"cs.AI","submitted_at":"2026-06-12T08:20:09+00:00","version_compiled":"v2"},"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 v2","sentences":[{"i":0,"p":0.99,"id":"s0","own":false,"role":"background","text":"Agent skills provide LLM agents with reusable procedures, but a given skill may be incomplete, ineffective, or misleading for the current task.","hedge":{"p":0.66,"label":"hedged_possibility"},"own_p":0.05,"main_claim_p":0.02,"quantitative":false,"quantitative_p":0.03,"role_probabilities":{"gap":0.01,"method":0.0,"result":0.0,"background":0.99,"limitation":0.0,"future_work":0.0,"contribution":0.0,"interpretation":0.0}},{"i":1,"p":0.98,"id":"s1","own":false,"role":"background","text":"Existing methods for revising skills typically rely on external outcome supervision, such as task rewards, performance on a separate validation set, or feedback from deployment, to decide which updates to retain.","hedge":{"p":1.0,"label":"asserted"},"own_p":0.03,"main_claim_p":0.02,"quantitative":false,"quantitative_p":0.04,"role_probabilities":{"gap":0.02,"method":0.0,"result":0.0,"background":0.98,"limitation":0.0,"future_work":0.0,"contribution":0.0,"interpretation":0.0}},{"i":2,"p":0.6,"id":"s2","own":false,"role":"background","text":"However, this supervision may be costly or unavailable when adapting a skill to a new task.","hedge":{"p":0.99,"label":"hedged_possibility"},"own_p":0.05,"main_claim_p":0.02,"quantitative":false,"quantitative_p":0.03,"role_probabilities":{"gap":0.38,"method":0.0,"result":0.0,"background":0.6,"limitation":0.02,"future_work":0.0,"contribution":0.0,"interpretation":0.0}},{"i":3,"p":0.61,"id":"s3","own":false,"role":"background","text":"Using execution behavior as an alternative signal is challenging: a single run reflects the combined effects of the agent, task, and workspace, making the skill's contribution difficult to isolate.","hedge":{"p":0.93,"label":"asserted"},"own_p":0.1,"main_claim_p":0.02,"quantitative":false,"quantitative_p":0.06,"role_probabilities":{"gap":0.37,"method":0.0,"result":0.0,"background":0.61,"limitation":0.02,"future_work":0.0,"contribution":0.0,"interpretation":0.0}},{"i":4,"p":0.67,"id":"s4","own":false,"role":"method","text":"Comparing executions with and without the skill, while holding these factors fixed, provides evidence about how the skill changes behavior.","hedge":{"p":0.87,"label":"asserted"},"own_p":0.15,"main_claim_p":0.06,"quantitative":false,"quantitative_p":0.24,"role_probabilities":{"gap":0.02,"method":0.67,"result":0.0,"background":0.28,"limitation":0.0,"future_work":0.0,"contribution":0.02,"interpretation":0.01}},{"i":5,"p":0.86,"id":"s5","own":true,"role":"contribution","text":"We introduce SelfEvoSkill, which links these behavioral differences to relevant skill passages and iteratively revises and reevaluates those passages through subsequent executions.","hedge":{"p":0.99,"label":"asserted"},"own_p":0.92,"main_claim_p":0.09,"quantitative":false,"quantitative_p":0.03,"role_probabilities":{"gap":0.0,"method":0.14,"result":0.0,"background":0.0,"limitation":0.0,"future_work":0.0,"contribution":0.86,"interpretation":0.0}},{"i":6,"p":0.96,"id":"s6","own":true,"role":"method","text":"Checks derived once from the task's explicit requirements prevent revisions from violating task constraints.","hedge":{"p":1.0,"label":"asserted"},"own_p":0.79,"main_claim_p":0.04,"quantitative":false,"quantitative_p":0.03,"role_probabilities":{"gap":0.0,"method":0.96,"result":0.0,"background":0.0,"limitation":0.0,"future_work":0.0,"contribution":0.04,"interpretation":0.0}},{"i":7,"p":1.0,"id":"s7","own":true,"role":"result","text":"Across 89 containerized tasks in 8 professional domains, under the same best-of-two evaluation budget, SelfEvoSkill attains a 73.9% mean task reward, exceeding the no-skill condition by +33.0 percentage points and the benchmark's curated skill by +17.2 percentage points, while skill evolution receives neither task rewards nor benchmark verifier outputs.","hedge":{"p":1.0,"label":"asserted"},"own_p":0.97,"main_claim_p":0.24,"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}},{"i":8,"p":1.0,"id":"s8","own":true,"role":"result","text":"In controlled ablations, using two with-skill runs instead of a matched with/without-skill pair reduces mean reward by 4.6 percentage points, while limiting revision to one round reduces it by 9.0 points relative to repeated revision.","hedge":{"p":1.0,"label":"asserted"},"own_p":0.96,"main_claim_p":0.06,"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}},{"i":9,"p":1.0,"id":"s9","own":true,"role":"interpretation","text":"Together, these results show that behavioral differences between matched executions can provide an effective signal for revising skills when outcome supervision is unavailable.","hedge":{"p":1.0,"label":"asserted"},"own_p":0.92,"main_claim_p":0.92,"quantitative":false,"quantitative_p":0.05,"role_probabilities":{"gap":0.0,"method":0.0,"result":0.0,"background":0.0,"limitation":0.0,"future_work":0.0,"contribution":0.0,"interpretation":1.0}}]},"compiled":{"main_claim":{"p":1.0,"text":"Together, these results show that behavioral differences between matched executions can provide an effective signal for revising skills when outcome supervision is unavailable.","location":"abstract s9","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":7726,"cost_usd_reported":0.0},"compiled_at":"2026-10-06","bank_version":"abstract-p1-20260923","model_version":"classifier-1.13.0","abstract_states":{"claims_first":0.08,"states_limitation":0.37,"human_participants":0.03,"controlled_comparison":0.97,"randomized_assignment":0.02,"states_quantitative_result":0.99,"compares_with_prior_results":0.97,"mentions_code_or_data_availability":0.03},"evidential_basis":{"p":0.95,"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":0.95,"simulation_or_computation":0.05}},"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":"We introduce SelfEvoSkill, which links these behavioral differences to relevant skill passages and iteratively revises and reevaluates those passages through subsequent executions.","location":"abstract s5"}},"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 v2"],"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"}}}