{"paper":{"title":"SkillNEO: Non-Elitist Evolutionary Optimization of Agent Skills","arxiv_id":"2610.20926","primary_cat":"cs.AI","submitted_at":"2026-10-14T20:57:26+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":"Agent skills turn procedural knowledge into reusable artifacts, allowing LLM agents to accumulate and refine expertise through experience.","hedge":{"p":1.0,"label":"asserted"},"own_p":0.06,"main_claim_p":0.03,"quantitative":false,"quantitative_p":0.02,"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":1.0,"id":"s1","own":false,"role":"background","text":"Building on this, skill evolution iteratively improves skills using execution feedback.","hedge":{"p":1.0,"label":"asserted"},"own_p":0.08,"main_claim_p":0.03,"quantitative":false,"quantitative_p":0.03,"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":2,"p":0.55,"id":"s2","own":false,"role":"background","text":"However, improvement may require several interdependent edits, so low-reward versions can still be worth developing.","hedge":{"p":1.0,"label":"hedged_possibility"},"own_p":0.1,"main_claim_p":0.04,"quantitative":false,"quantitative_p":0.07,"role_probabilities":{"gap":0.44,"method":0.0,"result":0.0,"background":0.55,"limitation":0.0,"future_work":0.0,"contribution":0.0,"interpretation":0.01}},{"i":3,"p":0.74,"id":"s3","own":false,"role":"background","text":"Requiring immediate reward gains can discard useful partial repairs before their benefits emerge.","hedge":{"p":0.61,"label":"hedged_possibility"},"own_p":0.08,"main_claim_p":0.04,"quantitative":false,"quantitative_p":0.04,"role_probabilities":{"gap":0.24,"method":0.0,"result":0.0,"background":0.74,"limitation":0.0,"future_work":0.0,"contribution":0.0,"interpretation":0.02}},{"i":4,"p":0.65,"id":"s4","own":false,"role":"method","text":"Non-elitist evolution allows new versions to receive further edits despite scoring below earlier ones.","hedge":{"p":0.99,"label":"asserted"},"own_p":0.17,"main_claim_p":0.06,"quantitative":false,"quantitative_p":0.09,"role_probabilities":{"gap":0.01,"method":0.65,"result":0.0,"background":0.22,"limitation":0.0,"future_work":0.0,"contribution":0.11,"interpretation":0.01}},{"i":5,"p":0.91,"id":"s5","own":false,"role":"gap","text":"For agent skills, this calls for evidence of what to preserve and repair next.","hedge":{"p":0.65,"label":"asserted"},"own_p":0.06,"main_claim_p":0.03,"quantitative":false,"quantitative_p":0.02,"role_probabilities":{"gap":0.91,"method":0.01,"result":0.0,"background":0.06,"limitation":0.0,"future_work":0.0,"contribution":0.02,"interpretation":0.0}},{"i":6,"p":0.96,"id":"s6","own":true,"role":"contribution","text":"We introduce SkillNEO (Non-Elitist Evolutionary Optimization for Agent Skills), which separately selects a version to use and versions to keep improving.","hedge":{"p":1.0,"label":"asserted"},"own_p":0.87,"main_claim_p":0.09,"quantitative":false,"quantitative_p":0.04,"role_probabilities":{"gap":0.0,"method":0.04,"result":0.0,"background":0.0,"limitation":0.0,"future_work":0.0,"contribution":0.96,"interpretation":0.0}},{"i":7,"p":0.97,"id":"s7","own":true,"role":"method","text":"Comparing how related versions perform the same task reveals working steps to preserve and remaining failures to fix.","hedge":{"p":0.57,"label":"asserted"},"own_p":0.5,"main_claim_p":0.06,"quantitative":false,"quantitative_p":0.04,"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":8,"p":0.83,"id":"s8","own":false,"role":"method","text":"It can also revisit and revise earlier versions using later evidence.","hedge":{"p":0.55,"label":"hedged_possibility"},"own_p":0.41,"main_claim_p":0.05,"quantitative":false,"quantitative_p":0.03,"role_probabilities":{"gap":0.0,"method":0.83,"result":0.0,"background":0.0,"limitation":0.0,"future_work":0.0,"contribution":0.16,"interpretation":0.01}},{"i":9,"p":1.0,"id":"s9","own":true,"role":"result","text":"Across 86 SkillsBench tasks in eight domains, including software engineering and office work, SkillNEO raises GPT-5.4 Mini's average reward from 49.1% to 66.4%, within GPT-5.5 and GPT-5.6 Luna's original-skill range.","hedge":{"p":1.0,"label":"asserted"},"own_p":0.96,"main_claim_p":0.25,"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":10,"p":1.0,"id":"s10","own":true,"role":"method","text":"Mini executes tasks, proposes revisions, and edits skills; final evaluation uses the same task instances.","hedge":{"p":0.99,"label":"asserted"},"own_p":0.9,"main_claim_p":0.03,"quantitative":false,"quantitative_p":0.08,"role_probabilities":{"gap":0.0,"method":1.0,"result":0.0,"background":0.0,"limitation":0.0,"future_work":0.0,"contribution":0.0,"interpretation":0.0}},{"i":11,"p":1.0,"id":"s11","own":true,"role":"result","text":"Gains span all eight domains.","hedge":{"p":1.0,"label":"asserted"},"own_p":0.92,"main_claim_p":0.04,"quantitative":false,"quantitative_p":0.44,"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":12,"p":0.99,"id":"s12","own":true,"role":"result","text":"The same skills improve stronger models without further editing.","hedge":{"p":1.0,"label":"asserted"},"own_p":0.88,"main_claim_p":0.11,"quantitative":false,"quantitative_p":0.09,"role_probabilities":{"gap":0.0,"method":0.0,"result":0.99,"background":0.0,"limitation":0.0,"future_work":0.0,"contribution":0.0,"interpretation":0.01}},{"i":13,"p":1.0,"id":"s13","own":true,"role":"result","text":"On the 41 tasks with modified skills, reuse improves Luna by 10.5 percentage points and places GPT-5.5 10.6 points above its published original-skill reference.","hedge":{"p":1.0,"label":"asserted"},"own_p":0.93,"main_claim_p":0.09,"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":14,"p":0.85,"id":"s14","own":true,"role":"result","text":"Execution records show these skills can benefit stronger models even when Mini still fails some task requirements.","hedge":{"p":0.74,"label":"asserted"},"own_p":0.92,"main_claim_p":0.09,"quantitative":false,"quantitative_p":0.07,"role_probabilities":{"gap":0.0,"method":0.0,"result":0.85,"background":0.0,"limitation":0.0,"future_work":0.0,"contribution":0.0,"interpretation":0.15}},{"i":15,"p":0.99,"id":"s15","own":true,"role":"result","text":"Search histories also show lower-scoring versions leading to skills ultimately selected for use.","hedge":{"p":1.0,"label":"asserted"},"own_p":0.94,"main_claim_p":0.06,"quantitative":false,"quantitative_p":0.13,"role_probabilities":{"gap":0.0,"method":0.0,"result":0.99,"background":0.0,"limitation":0.0,"future_work":0.0,"contribution":0.0,"interpretation":0.01}},{"i":16,"p":1.0,"id":"s16","own":true,"role":"interpretation","text":"Together, these findings show that skill evolution can turn a weaker model's partial progress into reusable procedural knowledge for improving more capable agents.","hedge":{"p":1.0,"label":"asserted"},"own_p":0.89,"main_claim_p":0.9,"quantitative":false,"quantitative_p":0.04,"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 findings show that skill evolution can turn a weaker model's partial progress into reusable procedural knowledge for improving more capable agents.","location":"abstract s16","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":11884,"cost_usd_reported":0.0},"compiled_at":"2026-10-16","bank_version":"abstract-p1-20260923","model_version":"classifier-1.13.0","abstract_states":{"claims_first":0.06,"states_limitation":0.61,"human_participants":0.04,"controlled_comparison":0.84,"randomized_assignment":0.02,"states_quantitative_result":0.99,"compares_with_prior_results":0.94,"mentions_code_or_data_availability":0.04},"evidential_basis":{"p":0.97,"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":0.97,"simulation_or_computation":0.03}},"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 SkillNEO (Non-Elitist Evolutionary Optimization for Agent Skills), which separately selects a version to use and versions to keep improving.","location":"abstract s6"}},"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":17,"sentences_in_abstract":17},"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"}}}