{"paper":{"title":"Aspire: Can Models Self-Evolve from Vague Goals?","arxiv_id":"2608.31111","primary_cat":"cs.CL","submitted_at":"2026-08-31T17:14:59+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":"Many important forms of human learning begin with a vague goal, such as \"become a better physicist\" or \"improve at research.\" Learners must interpret the goal, identify capability gaps, decide how to learn, and determine whether they have actually improved.","hedge":{"p":1.0,"label":"asserted"},"own_p":0.05,"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":0.65,"id":"s1","own":false,"role":"gap","text":"In contrast, existing work on LLM self-evolution typically begins with tasks and evaluation metrics specified by humans, reducing self-evolution to optimizing an explicit objective rather than deciding what and how to learn.","hedge":{"p":1.0,"label":"asserted"},"own_p":0.05,"main_claim_p":0.04,"quantitative":false,"quantitative_p":0.33,"role_probabilities":{"gap":0.65,"method":0.0,"result":0.0,"background":0.35,"limitation":0.0,"future_work":0.0,"contribution":0.0,"interpretation":0.0}},{"i":2,"p":1.0,"id":"s2","own":true,"role":"contribution","text":"We introduce ASPIRE, a benchmark for vague-goal-driven self-evolution.","hedge":{"p":0.98,"label":"asserted"},"own_p":0.85,"main_claim_p":0.03,"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":3,"p":0.71,"id":"s3","own":true,"role":"method","text":"ASPIRE provides only a natural-language capability goal while downstream evaluation tasks remain hidden.","hedge":{"p":1.0,"label":"asserted"},"own_p":0.73,"main_claim_p":0.03,"quantitative":false,"quantitative_p":0.05,"role_probabilities":{"gap":0.0,"method":0.71,"result":0.0,"background":0.0,"limitation":0.0,"future_work":0.0,"contribution":0.29,"interpretation":0.0}},{"i":4,"p":0.93,"id":"s4","own":false,"role":"method","text":"The agent must operationalize the goal by choosing data and update methods, constructing training and validation signals, and deciding when to evaluate.","hedge":{"p":0.99,"label":"asserted"},"own_p":0.23,"main_claim_p":0.05,"quantitative":false,"quantitative_p":0.03,"role_probabilities":{"gap":0.0,"method":0.93,"result":0.0,"background":0.0,"limitation":0.0,"future_work":0.0,"contribution":0.07,"interpretation":0.0}},{"i":5,"p":0.86,"id":"s5","own":true,"role":"method","text":"ASPIRE supports both model-weight and agent-harness evolution in a unified interactive environment and evaluates the resulting systems on a hidden, expert-authored set of 520 items spanning six goals.","hedge":{"p":1.0,"label":"asserted"},"own_p":0.9,"main_claim_p":0.05,"quantitative":true,"quantitative_p":0.98,"role_probabilities":{"gap":0.0,"method":0.86,"result":0.0,"background":0.0,"limitation":0.0,"future_work":0.0,"contribution":0.14,"interpretation":0.0}},{"i":6,"p":0.73,"id":"s6","own":true,"role":"result","text":"Our experiments show that vague goals redirect search effort toward goal interpretation.","hedge":{"p":1.0,"label":"asserted"},"own_p":0.93,"main_claim_p":0.24,"quantitative":false,"quantitative_p":0.06,"role_probabilities":{"gap":0.0,"method":0.0,"result":0.73,"background":0.0,"limitation":0.0,"future_work":0.0,"contribution":0.0,"interpretation":0.27}},{"i":7,"p":0.99,"id":"s7","own":true,"role":"result","text":"Current agents routinely complete training and harness-editing loops, but weight-level gains remain sparse and unstable, and the strongest evolved harness remains below the engineered Qwen-Agent reference.","hedge":{"p":1.0,"label":"asserted"},"own_p":0.91,"main_claim_p":0.41,"quantitative":true,"quantitative_p":0.96,"role_probabilities":{"gap":0.0,"method":0.0,"result":0.99,"background":0.0,"limitation":0.01,"future_work":0.0,"contribution":0.0,"interpretation":0.0}},{"i":8,"p":0.55,"id":"s8","own":true,"role":"interpretation","text":"Agents often train on mismatched data and trust narrow self-evaluations, so local gains fail to transfer to hidden evaluation and continued search and training can erase earlier improvements.","hedge":{"p":0.96,"label":"asserted"},"own_p":0.84,"main_claim_p":0.32,"quantitative":false,"quantitative_p":0.08,"role_probabilities":{"gap":0.0,"method":0.0,"result":0.44,"background":0.0,"limitation":0.01,"future_work":0.0,"contribution":0.0,"interpretation":0.55}}]},"compiled":{"main_claim":{"p":0.59,"text":"Current agents routinely complete training and harness-editing loops, but weight-level gains remain sparse and unstable, and the strongest evolved harness remains below the engineered Qwen-Agent reference.","location":"abstract s7","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":6949,"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":0.17,"states_limitation":0.89,"human_participants":0.09,"controlled_comparison":0.78,"randomized_assignment":0.02,"states_quantitative_result":0.91,"compares_with_prior_results":0.94,"mentions_code_or_data_availability":0.06},"evidential_basis":{"p":1.0,"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":1.0,"simulation_or_computation":0.0}},"contribution_type":{"p":1.0,"value":"dataset or benchmark","probabilities":{"review":0.0,"method_or_tool":0.0,"empirical_study":0.0,"simulation_study":0.0,"theoretical_result":0.0,"dataset_or_benchmark":1.0,"position_or_commentary":0.0}},"contribution_statement":{"p":0.99,"text":"We introduce ASPIRE, a benchmark for vague-goal-driven self-evolution.","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":9,"sentences_in_abstract":9},"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"}}}