{"paper":{"title":"TacitVLA: Reasoning Vision-Language-Action Models Without Reasoning Latency","arxiv_id":"2610.17540","primary_cat":"cs.AI","submitted_at":"2026-10-13T01:09:25+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":0.98,"id":"s0","own":false,"role":"background","text":"Textual reasoning before acting has improved vision-language-action (VLA) models, but at the cost of latency: every action conditioned on the reasoning text must wait for autoregressive decoding.","hedge":{"p":1.0,"label":"asserted"},"own_p":0.08,"main_claim_p":0.05,"quantitative":false,"quantitative_p":0.15,"role_probabilities":{"gap":0.02,"method":0.0,"result":0.0,"background":0.98,"limitation":0.0,"future_work":0.0,"contribution":0.0,"interpretation":0.0}},{"i":1,"p":0.4,"id":"s1","own":true,"role":"interpretation","text":"We show that actions do not need to wait for or condition on the generated reasoning to benefit from it, because the reasoning supervision already improves the backbone representation.","hedge":{"p":1.0,"label":"asserted"},"own_p":0.87,"main_claim_p":0.8,"quantitative":false,"quantitative_p":0.05,"role_probabilities":{"gap":0.01,"method":0.02,"result":0.2,"background":0.0,"limitation":0.0,"future_work":0.0,"contribution":0.37,"interpretation":0.4}},{"i":2,"p":0.57,"id":"s2","own":true,"role":"method","text":"We introduce TacitVLA, a reasoning VLA whose action expert never attends to the reasoning it generates for the current call.","hedge":{"p":1.0,"label":"asserted"},"own_p":0.84,"main_claim_p":0.19,"quantitative":false,"quantitative_p":0.04,"role_probabilities":{"gap":0.0,"method":0.57,"result":0.0,"background":0.0,"limitation":0.0,"future_work":0.0,"contribution":0.43,"interpretation":0.0}},{"i":3,"p":0.87,"id":"s3","own":true,"role":"method","text":"On long-horizon tasks where the reasoning text also serves as episodic memory, it can generate each entry during action execution rather than before acting.","hedge":{"p":0.79,"label":"hedged_possibility"},"own_p":0.64,"main_claim_p":0.08,"quantitative":false,"quantitative_p":0.04,"role_probabilities":{"gap":0.0,"method":0.87,"result":0.0,"background":0.0,"limitation":0.0,"future_work":0.0,"contribution":0.13,"interpretation":0.0}},{"i":4,"p":0.7,"id":"s4","own":true,"role":"method","text":"We evaluate it in a controlled study on four benchmarks spanning manipulation and autonomous driving, memory and non-memory tasks: RoboMME, RMBench, LIBERO, and PhysicalAI-AV.","hedge":{"p":0.96,"label":"asserted"},"own_p":0.93,"main_claim_p":0.03,"quantitative":true,"quantitative_p":0.86,"role_probabilities":{"gap":0.0,"method":0.7,"result":0.0,"background":0.0,"limitation":0.0,"future_work":0.0,"contribution":0.3,"interpretation":0.0}},{"i":5,"p":1.0,"id":"s5","own":true,"role":"result","text":"On all four, TacitVLA matches the performance of its explicit-reasoning counterpart at up to 3.4x lower latency, and matches the latency of the same architecture trained without reasoning at up to +17.1% success.","hedge":{"p":1.0,"label":"asserted"},"own_p":0.96,"main_claim_p":0.5,"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":6,"p":1.0,"id":"s6","own":true,"role":"result","text":"In absolute terms, it surpasses the best previously published results on the RoboMME Counting suite (82.7% vs. 66.8%) and RMBench (77.5% vs. 56.5%), and reaches 72.9% on LIBERO-Plus and a 52.5% scene score on the public split of the PhysicalAI-AV closed-loop driving challenge.","hedge":{"p":1.0,"label":"asserted"},"own_p":0.91,"main_claim_p":0.22,"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":7,"p":0.56,"id":"s7","own":false,"role":"contribution","text":"Project page: https://tacitvla.github.io.","hedge":{"p":0.98,"label":"unclear"},"own_p":0.1,"main_claim_p":0.01,"quantitative":false,"quantitative_p":0.02,"role_probabilities":{"gap":0.0,"method":0.04,"result":0.02,"background":0.13,"limitation":0.0,"future_work":0.25,"contribution":0.56,"interpretation":0.0}}]},"compiled":{"main_claim":{"p":0.88,"text":"We show that actions do not need to wait for or condition on the generated reasoning to benefit from it, because the reasoning supervision already improves the backbone representation.","location":"abstract s1","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":6499,"cost_usd_reported":0.0},"compiled_at":"2026-10-14","bank_version":"abstract-p1-20260923","model_version":"classifier-1.13.0","abstract_states":{"claims_first":0.07,"states_limitation":0.17,"human_participants":0.05,"controlled_comparison":0.95,"randomized_assignment":0.02,"states_quantitative_result":0.99,"compares_with_prior_results":0.98,"mentions_code_or_data_availability":0.34},"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":0.99,"text":"We introduce TacitVLA, a reasoning VLA whose action expert never attends to the reasoning it generates for the current call.","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":8,"sentences_in_abstract":8},"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"}}}