{"paper":{"title":"Blitz: Accelerating the Diffusion Modules of AlphaFold3 and Boltz-2","arxiv_id":"2610.23703","primary_cat":"cs.LG","submitted_at":"2026-10-16T16:29:04+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":"Diffusion-based protein structure prediction models produce all-atom structures in about 200 denoising steps, and that cost has driven a number of few-step samplers.","hedge":{"p":1.0,"label":"asserted"},"own_p":0.09,"main_claim_p":0.03,"quantitative":true,"quantitative_p":0.98,"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":0.58,"id":"s1","own":true,"role":"method","text":"We present Blitz, a few-step distillation framework that uses a geometric noise schedule to distribute steps more evenly across noise scales, then learns those updates through multistep consistency training.","hedge":{"p":1.0,"label":"asserted"},"own_p":0.93,"main_claim_p":0.07,"quantitative":false,"quantitative_p":0.05,"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":2,"p":0.95,"id":"s2","own":true,"role":"method","text":"Physical steering and target-free candidate selection help reduce the remaining geometric defects at inference.","hedge":{"p":0.92,"label":"asserted"},"own_p":0.74,"main_claim_p":0.04,"quantitative":false,"quantitative_p":0.03,"role_probabilities":{"gap":0.0,"method":0.95,"result":0.0,"background":0.0,"limitation":0.0,"future_work":0.0,"contribution":0.05,"interpretation":0.0}},{"i":3,"p":0.65,"id":"s3","own":true,"role":"result","text":"We apply Blitz to AlphaFold 3 and Boltz-2, reducing denoising from 200 steps to eight.","hedge":{"p":1.0,"label":"asserted"},"own_p":0.96,"main_claim_p":0.25,"quantitative":true,"quantitative_p":0.99,"role_probabilities":{"gap":0.0,"method":0.02,"result":0.65,"background":0.0,"limitation":0.0,"future_work":0.0,"contribution":0.33,"interpretation":0.0}},{"i":4,"p":0.62,"id":"s4","own":true,"role":"method","text":"Because published few-step results use different benchmarks and scoring procedures, we compare both distilled models with their teachers and other few-step methods under a common evaluation protocol covering structural accuracy, interface quality, ligand placement, and stereochemical validity.","hedge":{"p":0.94,"label":"asserted"},"own_p":0.9,"main_claim_p":0.05,"quantitative":true,"quantitative_p":0.5,"role_probabilities":{"gap":0.15,"method":0.62,"result":0.0,"background":0.01,"limitation":0.0,"future_work":0.0,"contribution":0.22,"interpretation":0.0}},{"i":5,"p":1.0,"id":"s5","own":true,"role":"result","text":"Blitz retains teacher-level structural accuracy, with AF3-Blitz outperforming all evaluated few-step baselines and exceeding its teacher on several accuracy metrics.","hedge":{"p":1.0,"label":"asserted"},"own_p":0.95,"main_claim_p":0.82,"quantitative":true,"quantitative_p":0.96,"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}}]},"compiled":{"main_claim":{"p":1.0,"text":"Blitz retains teacher-level structural accuracy, with AF3-Blitz outperforming all evaluated few-step baselines and exceeding its teacher on several accuracy metrics.","location":"abstract s5","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":4985,"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":0.04,"states_limitation":0.16,"human_participants":0.03,"controlled_comparison":0.95,"randomized_assignment":0.01,"states_quantitative_result":0.98,"compares_with_prior_results":0.98,"mentions_code_or_data_availability":0.04},"evidential_basis":{"p":0.7,"value":"simulation or computation","probabilities":{"argument":0.0,"literature":0.0,"proof_or_derivation":0.0,"analysis_of_existing_data":0.01,"observation_or_experiment":0.29,"simulation_or_computation":0.7}},"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":1.0,"text":"We present Blitz, a few-step distillation framework that uses a geometric noise schedule to distribute steps more evenly across noise scales, then learns those updates through multistep consistency training.","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":6,"sentences_in_abstract":6},"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}}