{"paper":{"title":"Evo$^2$: Evolving the Evolver through Weight-Harness Co-Evolution","arxiv_id":"2610.13025","primary_cat":"cs.CL","submitted_at":"2026-10-09T05:28:02+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":"Self-evolving agents are increasingly used to improve language model systems without human intervention through either weight updates or harness improvements.","hedge":{"p":0.99,"label":"asserted"},"own_p":0.04,"main_claim_p":0.02,"quantitative":false,"quantitative_p":0.07,"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":0.96,"id":"s1","own":false,"role":"background","text":"Existing approaches typically pursue these directions separately: some methods update model weights using self-generated experience, while others rewrite the harness that defines how an agent uses prompts, tools, context, and execution feedback.","hedge":{"p":0.99,"label":"asserted"},"own_p":0.04,"main_claim_p":0.02,"quantitative":false,"quantitative_p":0.21,"role_probabilities":{"gap":0.04,"method":0.0,"result":0.0,"background":0.96,"limitation":0.0,"future_work":0.0,"contribution":0.0,"interpretation":0.0}},{"i":2,"p":1.0,"id":"s2","own":false,"role":"gap","text":"However, these approaches overlook the relationship between model weights and the agent harness.","hedge":{"p":1.0,"label":"asserted"},"own_p":0.09,"main_claim_p":0.04,"quantitative":false,"quantitative_p":0.02,"role_probabilities":{"gap":1.0,"method":0.0,"result":0.0,"background":0.0,"limitation":0.0,"future_work":0.0,"contribution":0.0,"interpretation":0.0}},{"i":3,"p":1.0,"id":"s3","own":true,"role":"contribution","text":"To connect these processes and ensure effective and smooth agent self-evolving, we propose Evo$^2$, a framework for weight-harness co-evolution.","hedge":{"p":0.94,"label":"asserted"},"own_p":0.73,"main_claim_p":0.09,"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":4,"p":0.96,"id":"s4","own":true,"role":"method","text":"For harness updates, we design a unified dual-role pipeline that a single policy serves as both a Proposer, which writes harness edits, and a Reflector, which diagnoses their effects and guides revision.","hedge":{"p":1.0,"label":"asserted"},"own_p":0.92,"main_claim_p":0.04,"quantitative":false,"quantitative_p":0.11,"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":5,"p":0.99,"id":"s5","own":true,"role":"method","text":"For weight updates, harness modifications and their execution outcomes guide the training of both the Proposer and the Reflector.","hedge":{"p":1.0,"label":"asserted"},"own_p":0.84,"main_claim_p":0.04,"quantitative":false,"quantitative_p":0.02,"role_probabilities":{"gap":0.0,"method":0.99,"result":0.0,"background":0.0,"limitation":0.0,"future_work":0.0,"contribution":0.01,"interpretation":0.0}},{"i":6,"p":0.98,"id":"s6","own":true,"role":"method","text":"We further develop EvoTIDE, a training pipeline that recovers usable trajectories through edit extraction and prefix continuation, then combines SFT with interleaved GRPO across the two roles.","hedge":{"p":1.0,"label":"asserted"},"own_p":0.94,"main_claim_p":0.03,"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":0.99,"id":"s7","own":true,"role":"result","text":"Extensive experiments across three coding benchmarks demonstrate that Evo$^2$ significantly outperforms existing self-evolution methods.","hedge":{"p":1.0,"label":"asserted"},"own_p":0.96,"main_claim_p":0.61,"quantitative":true,"quantitative_p":0.59,"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":8,"p":1.0,"id":"s8","own":true,"role":"result","text":"It achieves 82.1% on Polyglot and 48.3% on SWE-bench Verified, compared with 62.6% and 41.0% for the strongest baseline, respectively.","hedge":{"p":1.0,"label":"asserted"},"own_p":0.95,"main_claim_p":0.15,"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":"result","text":"We also show that the evolved harnesses transfer to different tasks without further evolution and rank first in cross-benchmark evaluations.","hedge":{"p":1.0,"label":"asserted"},"own_p":0.96,"main_claim_p":0.16,"quantitative":true,"quantitative_p":0.72,"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":0.99,"id":"s10","own":true,"role":"result","text":"Our dual-role harness improvement framework is also able to extend to self-evolution with proprietary models, improving a GPT-5-mini agent's performance from 24.4% to 48.3%.","hedge":{"p":1.0,"label":"asserted"},"own_p":0.95,"main_claim_p":0.08,"quantitative":true,"quantitative_p":0.99,"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":11,"p":1.0,"id":"s11","own":true,"role":"interpretation","text":"These results demonstrate the applicability of the framework Evo$^2$ across domains and its broad transferability across models and tasks.","hedge":{"p":1.0,"label":"asserted"},"own_p":0.88,"main_claim_p":0.47,"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":0.0,"interpretation":1.0}}]},"compiled":{"main_claim":{"p":0.56,"text":"These results demonstrate the applicability of the framework Evo$^2$ across domains and its broad transferability across models and tasks.","location":"abstract s11","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":8996,"cost_usd_reported":0.0},"compiled_at":"2026-10-12","bank_version":"abstract-p1-20260923","model_version":"classifier-1.13.0","abstract_states":{"claims_first":0.1,"states_limitation":0.08,"human_participants":0.04,"controlled_comparison":0.95,"randomized_assignment":0.01,"states_quantitative_result":0.99,"compares_with_prior_results":0.98,"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":0.98,"text":"To connect these processes and ensure effective and smooth agent self-evolving, we propose Evo$^2$, a framework for weight-harness co-evolution.","location":"abstract s3"}},"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":12,"sentences_in_abstract":12},"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"}}}