{"paper":{"title":"Trustworthy Recursive Self-Improvement in the Era of Large Models: Foundation, Risks, Evaluation and Open Challenges","arxiv_id":"2610.35193","primary_cat":"cs.AI","submitted_at":"2026-10-26T10:46:03+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":"AI systems built around large models increasingly use experience and feedback to make persistent changes to models, prompts, memories, tools, and workflows.","hedge":{"p":0.99,"label":"asserted"},"own_p":0.07,"main_claim_p":0.03,"quantitative":false,"quantitative_p":0.04,"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":"Recursive self-improvement (RSI) extends this process to inductive biases and to the mechanisms that govern future updates, which can themselves be revised and reused across improvement cycles.","hedge":{"p":1.0,"label":"asserted"},"own_p":0.07,"main_claim_p":0.05,"quantitative":false,"quantitative_p":0.04,"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":0.72,"id":"s2","own":false,"role":"background","text":"Yet self-modification alone does not guarantee trustworthy improvement.","hedge":{"p":1.0,"label":"asserted"},"own_p":0.12,"main_claim_p":0.23,"quantitative":false,"quantitative_p":0.05,"role_probabilities":{"gap":0.25,"method":0.0,"result":0.0,"background":0.72,"limitation":0.02,"future_work":0.0,"contribution":0.0,"interpretation":0.01}},{"i":3,"p":0.92,"id":"s3","own":false,"role":"background","text":"Reported gains may reflect flawed evaluations, previously acquired capabilities may deteriorate, and compliance with behavioral constraints may weaken.","hedge":{"p":0.98,"label":"hedged_possibility"},"own_p":0.09,"main_claim_p":0.1,"quantitative":false,"quantitative_p":0.04,"role_probabilities":{"gap":0.05,"method":0.0,"result":0.01,"background":0.92,"limitation":0.02,"future_work":0.0,"contribution":0.0,"interpretation":0.0}},{"i":4,"p":0.95,"id":"s4","own":false,"role":"background","text":"Compromised artifacts or sensitive information may also persist in successor systems.","hedge":{"p":0.99,"label":"hedged_possibility"},"own_p":0.06,"main_claim_p":0.04,"quantitative":false,"quantitative_p":0.03,"role_probabilities":{"gap":0.02,"method":0.0,"result":0.01,"background":0.95,"limitation":0.02,"future_work":0.0,"contribution":0.0,"interpretation":0.0}},{"i":5,"p":0.95,"id":"s5","own":false,"role":"background","text":"Changes to improvement mechanisms may invalidate earlier verification results and make human oversight less effective.","hedge":{"p":0.75,"label":"hedged_possibility"},"own_p":0.09,"main_claim_p":0.07,"quantitative":false,"quantitative_p":0.05,"role_probabilities":{"gap":0.02,"method":0.0,"result":0.0,"background":0.95,"limitation":0.02,"future_work":0.0,"contribution":0.0,"interpretation":0.01}},{"i":6,"p":1.0,"id":"s6","own":true,"role":"contribution","text":"This survey presents an operational definition and process framework for RSI and organizes the requirements for trustworthy RSI into four interconnected themes: genuine and sustained improvement; behavioral constraints, alignment, and fairness; system security and privacy; and transparency and controllability.","hedge":{"p":0.98,"label":"asserted"},"own_p":0.89,"main_claim_p":0.18,"quantitative":true,"quantitative_p":0.74,"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":7,"p":0.52,"id":"s7","own":true,"role":"method","text":"We examine these themes from three perspectives: Improvement Validity, Constraint Preservation, and Effective Oversight.","hedge":{"p":0.93,"label":"asserted"},"own_p":0.76,"main_claim_p":0.06,"quantitative":false,"quantitative_p":0.45,"role_probabilities":{"gap":0.0,"method":0.52,"result":0.0,"background":0.0,"limitation":0.0,"future_work":0.0,"contribution":0.48,"interpretation":0.0}},{"i":8,"p":0.58,"id":"s8","own":true,"role":"method","text":"This analysis covers improvement diagnosis, proposal generation, proposal evaluation, and update implementation (S1--S4), including the recursive application of these stages to improvement mechanisms (RI).","hedge":{"p":0.98,"label":"asserted"},"own_p":0.75,"main_claim_p":0.05,"quantitative":true,"quantitative_p":0.51,"role_probabilities":{"gap":0.0,"method":0.58,"result":0.0,"background":0.0,"limitation":0.0,"future_work":0.0,"contribution":0.42,"interpretation":0.0}},{"i":9,"p":0.83,"id":"s9","own":true,"role":"contribution","text":"We review risks, safeguards, and evaluation evidence for individual updates, effects inherited by successor systems, and revisions to improvement mechanisms, while distinguishing direct RSI findings from evidence in related systems.","hedge":{"p":0.91,"label":"asserted"},"own_p":0.61,"main_claim_p":0.08,"quantitative":false,"quantitative_p":0.16,"role_probabilities":{"gap":0.0,"method":0.17,"result":0.0,"background":0.0,"limitation":0.0,"future_work":0.0,"contribution":0.83,"interpretation":0.0}},{"i":10,"p":0.7,"id":"s10","own":true,"role":"future_work","text":"Finally, we examine evaluation settings, identify requirements for independent and longitudinal assessment, and discuss applications and open research problems.","hedge":{"p":0.76,"label":"asserted"},"own_p":0.79,"main_claim_p":0.08,"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":0.7,"contribution":0.3,"interpretation":0.0}}]},"compiled":{"main_claim":{"p":0.88,"text":"Yet self-modification alone does not guarantee trustworthy improvement.","location":"abstract s2","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":8153,"cost_usd_reported":0.0},"compiled_at":"2026-10-27","bank_version":"abstract-p1-20260923","model_version":"classifier-1.13.0","abstract_states":{"claims_first":0.04,"states_limitation":0.32,"human_participants":0.08,"controlled_comparison":0.04,"randomized_assignment":0.01,"states_quantitative_result":0.03,"compares_with_prior_results":0.09,"mentions_code_or_data_availability":0.02},"evidential_basis":{"p":0.99,"value":"literature","probabilities":{"argument":0.01,"literature":0.99,"proof_or_derivation":0.0,"analysis_of_existing_data":0.0,"observation_or_experiment":0.0,"simulation_or_computation":0.0}},"contribution_type":{"p":1.0,"value":"review","probabilities":{"review":1.0,"method_or_tool":0.0,"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":"This survey presents an operational definition and process framework for RSI and organizes the requirements for trustworthy RSI into four interconnected themes: genuine and sustained improvement; behavioral constraints, alignment, and fairness; system security and privacy; and transparency and controllability.","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":11,"sentences_in_abstract":11},"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"}}}