{"paper":{"title":"Where is Physical RSI capability from? An Observational Scaling Law of LLM-as-policy","arxiv_id":"2610.26848","primary_cat":"cs.RO","submitted_at":"2026-10-19T17:59:30+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":0.97,"id":"s0","own":true,"role":"contribution","text":"In this paper, we systematically investigate the mechanisms, scaling behavior, and generalization of physical Recursive self-improvement.","hedge":{"p":0.65,"label":"asserted"},"own_p":0.76,"main_claim_p":0.05,"quantitative":false,"quantitative_p":0.04,"role_probabilities":{"gap":0.0,"method":0.0,"result":0.0,"background":0.03,"limitation":0.0,"future_work":0.0,"contribution":0.97,"interpretation":0.0}},{"i":1,"p":0.61,"id":"s1","own":true,"role":"method","text":"We introduce PhysRSI-eval to quantify physical RSI through a five-axis metric: First-Attempt Performance, Efficiency, Success, Budget and Improvement from failure.","hedge":{"p":1.0,"label":"asserted"},"own_p":0.96,"main_claim_p":0.05,"quantitative":true,"quantitative_p":0.8,"role_probabilities":{"gap":0.0,"method":0.61,"result":0.0,"background":0.0,"limitation":0.0,"future_work":0.0,"contribution":0.39,"interpretation":0.0}},{"i":2,"p":0.98,"id":"s2","own":true,"role":"method","text":"Different VLMs are evaluated on 18 manipulation tasks under a fixed interaction budget.","hedge":{"p":0.98,"label":"asserted"},"own_p":0.93,"main_claim_p":0.03,"quantitative":true,"quantitative_p":0.97,"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":3,"p":0.96,"id":"s3","own":true,"role":"interpretation","text":"The analysis of experimental results motivates the Physical Barrel Hypothesis: different capabilities may constrain different parts of improvement.","hedge":{"p":0.93,"label":"hedged_possibility"},"own_p":0.83,"main_claim_p":0.32,"quantitative":false,"quantitative_p":0.06,"role_probabilities":{"gap":0.0,"method":0.0,"result":0.02,"background":0.0,"limitation":0.0,"future_work":0.0,"contribution":0.02,"interpretation":0.96}},{"i":4,"p":0.89,"id":"s4","own":true,"role":"result","text":"The five-axis metrics also distinguish systems with close public benchmark averages and show that high scores need not come with early or efficient gains.","hedge":{"p":1.0,"label":"asserted"},"own_p":0.93,"main_claim_p":0.13,"quantitative":true,"quantitative_p":0.54,"role_probabilities":{"gap":0.0,"method":0.0,"result":0.89,"background":0.0,"limitation":0.0,"future_work":0.0,"contribution":0.01,"interpretation":0.1}},{"i":5,"p":0.81,"id":"s5","own":true,"role":"contribution","text":"We study observational scaling from two perspectives: predicting the relative performance of VLM agents from their public capabilities, and forecasting later progress from early physical feedback.","hedge":{"p":0.91,"label":"asserted"},"own_p":0.89,"main_claim_p":0.06,"quantitative":true,"quantitative_p":0.54,"role_probabilities":{"gap":0.0,"method":0.19,"result":0.0,"background":0.0,"limitation":0.0,"future_work":0.0,"contribution":0.81,"interpretation":0.0}},{"i":6,"p":0.52,"id":"s6","own":false,"role":"interpretation","text":"Together, these analyses ask how quickly and how far an agent can improve.","hedge":{"p":0.63,"label":"unclear"},"own_p":0.25,"main_claim_p":0.08,"quantitative":false,"quantitative_p":0.06,"role_probabilities":{"gap":0.08,"method":0.01,"result":0.0,"background":0.17,"limitation":0.0,"future_work":0.01,"contribution":0.21,"interpretation":0.52}},{"i":7,"p":0.51,"id":"s7","own":true,"role":"method","text":"Moreover, we conduct workspace interventions and then analyze what improvement leaves behind.","hedge":{"p":0.93,"label":"asserted"},"own_p":0.94,"main_claim_p":0.05,"quantitative":false,"quantitative_p":0.05,"role_probabilities":{"gap":0.0,"method":0.51,"result":0.0,"background":0.0,"limitation":0.0,"future_work":0.0,"contribution":0.49,"interpretation":0.0}},{"i":8,"p":1.0,"id":"s8","own":true,"role":"result","text":"Removing history reduces late-round success, frozen workspaces remain useful on unseen layouts, and three cross-task cases achieve success earlier.","hedge":{"p":1.0,"label":"asserted"},"own_p":0.94,"main_claim_p":0.11,"quantitative":true,"quantitative_p":0.85,"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":"interpretation","text":"The results suggest that scene-specific adaptation and reusable experience can coexist.","hedge":{"p":1.0,"label":"hedged_possibility"},"own_p":0.89,"main_claim_p":0.32,"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":0.0,"contribution":0.0,"interpretation":1.0}}]},"compiled":{"main_claim":{"p":0.77,"text":"The results suggest that scene-specific adaptation and reusable experience can coexist.","location":"abstract s9","hedge_strength":{"p":1.0,"label":"hedged possibility"}},"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":7357,"cost_usd_reported":0.0},"compiled_at":"2026-10-20","bank_version":"abstract-p1-20260923","model_version":"classifier-1.13.0","abstract_states":{"claims_first":0.06,"states_limitation":0.21,"human_participants":0.04,"controlled_comparison":0.42,"randomized_assignment":0.02,"states_quantitative_result":0.56,"compares_with_prior_results":0.51,"mentions_code_or_data_availability":0.04},"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":0.98,"value":"empirical study","probabilities":{"review":0.0,"method_or_tool":0.01,"empirical_study":0.98,"simulation_study":0.0,"theoretical_result":0.0,"dataset_or_benchmark":0.01,"position_or_commentary":0.0}},"contribution_statement":{"p":0.92,"text":"We introduce PhysRSI-eval to quantify physical RSI through a five-axis metric: First-Attempt Performance, Efficiency, Success, Budget and Improvement from failure.","location":"abstract s1"}},"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":10,"sentences_in_abstract":10},"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"}}}