{"paper":{"title":"SkillGLoW: Procedural-Family Skill Consolidation for Self-Improving Agents on Long-Horizon Task Streams","arxiv_id":"2609.02217","primary_cat":"cs.AI","submitted_at":"2026-09-02T07:31:18+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":"LLM agents increasingly self-improve by writing and reusing textual skills, kept either as one global document or as a flat pool of per-task entries, though most of the evidence comes from domains with structurally similar tasks.","hedge":{"p":0.97,"label":"asserted"},"own_p":0.06,"main_claim_p":0.03,"quantitative":false,"quantitative_p":0.15,"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":0.6,"id":"s1","own":false,"role":"gap","text":"On long-horizon workloads where each task demands a different solution, the two forms fail in opposite ways: the document collapses into generic discipline, while the pool inflates and its entries stay bound to the instance that wrote them.","hedge":{"p":1.0,"label":"asserted"},"own_p":0.2,"main_claim_p":0.08,"quantitative":false,"quantitative_p":0.45,"role_probabilities":{"gap":0.6,"method":0.0,"result":0.0,"background":0.4,"limitation":0.0,"future_work":0.0,"contribution":0.0,"interpretation":0.0}},{"i":2,"p":0.8,"id":"s2","own":true,"role":"method","text":"We argue the missing unit of reuse is the solving procedure shared by a cluster of related tasks, and build SkillGLoW (Global-Local Weave) around it: the local skills a task writes from its own execution are aggregated into procedural families and compressed into de-instantiated global priors, while the instance detail they hold is regenerated per task rather than stored; a commit gate admits a prior only when real execution shows it does not degrade the deployed library.","hedge":{"p":0.94,"label":"asserted"},"own_p":0.89,"main_claim_p":0.45,"quantitative":false,"quantitative_p":0.06,"role_probabilities":{"gap":0.0,"method":0.8,"result":0.0,"background":0.0,"limitation":0.0,"future_work":0.0,"contribution":0.2,"interpretation":0.0}},{"i":3,"p":1.0,"id":"s3","own":true,"role":"result","text":"Across four benchmarks (mathematical reasoning, terminal automation, software repair, and embodied control) and three models, the priors gain 17.2 points (hard) over the no-skill baseline on average, with positive gains in all 12 continual-improvement runs, and 18.0 with local regeneration, while the library holds one prior per procedural family, 3.6x more compact than the per-task pool.","hedge":{"p":1.0,"label":"asserted"},"own_p":0.96,"main_claim_p":0.37,"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":4,"p":1.0,"id":"s4","own":true,"role":"result","text":"Under the same protocol GLoW leads a published single-document optimizer on 15 of 21 cells.","hedge":{"p":1.0,"label":"asserted"},"own_p":0.87,"main_claim_p":0.09,"quantitative":true,"quantitative_p":0.98,"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":5,"p":0.64,"id":"s5","own":true,"role":"interpretation","text":"Unmodified, the library lifts success on unseen ALFWorld tasks from 73.9% to 83.9%, evidence that what transfers is procedure rather than task memory.","hedge":{"p":0.92,"label":"asserted"},"own_p":0.92,"main_claim_p":0.18,"quantitative":true,"quantitative_p":0.99,"role_probabilities":{"gap":0.0,"method":0.0,"result":0.36,"background":0.0,"limitation":0.0,"future_work":0.0,"contribution":0.0,"interpretation":0.64}}]},"compiled":{"main_claim":{"p":0.4,"text":"Unmodified, the library lifts success on unseen ALFWorld tasks from 73.9% to 83.9%, evidence that what transfers is procedure rather than task memory.","location":"abstract s5","hedge_strength":{"p":0.92,"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":5329,"cost_usd_reported":0.0},"compiled_at":"2026-09-23","bank_version":"abstract-p1-20260923","model_version":"classifier-1.13.0","abstract_states":{"claims_first":0.08,"states_limitation":0.25,"human_participants":0.03,"controlled_comparison":0.97,"randomized_assignment":0.02,"states_quantitative_result":0.99,"compares_with_prior_results":0.98,"mentions_code_or_data_availability":0.03},"evidential_basis":{"p":0.96,"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":0.96,"simulation_or_computation":0.04}},"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 argue the missing unit of reuse is the solving procedure shared by a cluster of related tasks, and build SkillGLoW (Global-Local Weave) around it: the local skills a task writes from its own execution are aggregated into procedural families and compressed into de-instantiated global priors, while the instance detail they hold is regenerated per task rather than stored; a commit gate admits a prior only when real execution shows it does not degrade the deployed library.","location":"abstract s2"}},"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":6,"sentences_in_abstract":6},"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"}}}