{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:FLL3MXXNNADKFGBIVLKXUB5UQC","merge_version":"pith-open-graph-merge-v1","event_count":2,"valid_event_count":2,"invalid_event_count":0,"equivocation_count":0,"current":{"canonical_record":{"metadata":{"abstract_canon_sha256":"5654583cd71b041baba26beae7e5844482c33556f0cc52ddf81e0b4dc4d3ae51","cross_cats_sorted":["cs.LG"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.RO","submitted_at":"2024-07-11T17:31:01Z","title_canon_sha256":"ab4d44c4fd049cf8a78071f32caa91ce066d78cd9a1a1a36d75e8876fdf214b1"},"schema_version":"1.0","source":{"id":"2407.08693","kind":"arxiv","version":3}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2407.08693","created_at":"2026-05-17T23:38:53Z"},{"alias_kind":"arxiv_version","alias_value":"2407.08693v3","created_at":"2026-05-17T23:38:53Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2407.08693","created_at":"2026-05-17T23:38:53Z"},{"alias_kind":"pith_short_12","alias_value":"FLL3MXXNNADK","created_at":"2026-05-18T12:33:37Z"},{"alias_kind":"pith_short_16","alias_value":"FLL3MXXNNADKFGBI","created_at":"2026-05-18T12:33:37Z"},{"alias_kind":"pith_short_8","alias_value":"FLL3MXXN","created_at":"2026-05-18T12:33:37Z"}],"graph_snapshots":[{"event_id":"sha256:82fd301cb652a6d54eb6c272900f074729ce007904d2fdc9473b852ffe359aae","target":"graph","created_at":"2026-05-17T23:38:53Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"graph_snapshot":{"author_claims":{"count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","strong_count":0},"builder_version":"pith-number-builder-2026-05-17-v1","claims":{"count":4,"items":[{"attestation":"unclaimed","claim_id":"C1","kind":"strongest_claim","source":"verdict.strongest_claim","status":"machine_extracted","text":"ECoT increases the absolute success rate of OpenVLA, the current strongest open-source VLA policy, by 28% across challenging generalization tasks, without any additional robot training data."},{"attestation":"unclaimed","claim_id":"C2","kind":"weakest_assumption","source":"verdict.weakest_assumption","status":"machine_extracted","text":"The synthetic data generation pipeline produces reasoning traces that are both accurate enough to supervise the model and sufficiently diverse to improve generalization rather than overfitting to the generation heuristics."},{"attestation":"unclaimed","claim_id":"C3","kind":"one_line_summary","source":"verdict.one_line_summary","status":"machine_extracted","text":"Training VLAs to perform embodied chain-of-thought reasoning about plans, sub-tasks, motions, and grounded visual features before acting raises OpenVLA success rates by 28% on challenging generalization tasks without new robot data."},{"attestation":"unclaimed","claim_id":"C4","kind":"headline","source":"verdict.pith_extraction.headline","status":"machine_extracted","text":"Embodied chain-of-thought reasoning trains VLAs to output grounded plans and visuals before actions, raising OpenVLA success by 28 percent."}],"snapshot_sha256":"1e95f37dfcfa3fd3139ba5bfdf7503823af107ebeabbb176dd60e81bf771352a"},"formal_canon":{"evidence_count":2,"snapshot_sha256":"41b360fcccc4df08dd6d710018e583705ec1716c285131c5f02ea538855243a6"},"paper":{"abstract_excerpt":"A key limitation of learned robot control policies is their inability to generalize outside their training data. Recent works on vision-language-action models (VLAs) have shown that the use of large, internet pre-trained vision-language models as the backbone of learned robot policies can substantially improve their robustness and generalization ability. Yet, one of the most exciting capabilities of large vision-language models in other domains is their ability to reason iteratively through complex problems. Can that same capability be brought into robotics to allow policies to improve perform","authors_text":"Chelsea Finn, Karl Pertsch, Micha{\\l} Zawalski, Oier Mees, Sergey Levine, William Chen","cross_cats":["cs.LG"],"headline":"Embodied chain-of-thought reasoning trains VLAs to output grounded plans and visuals before actions, raising OpenVLA success by 28 percent.","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.RO","submitted_at":"2024-07-11T17:31:01Z","title":"Robotic Control via Embodied Chain-of-Thought Reasoning"},"references":{"count":118,"internal_anchors":2,"resolved_work":118,"sample":[{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":1,"title":"A. Agarwal, A. Kumar, J. Malik, and D. Pathak. Legged locomotion in challenging terrains using egocentric vision, 2022","work_id":"b341c505-ba1d-4447-ae59-f044b8548c55","year":2022},{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":2,"title":"T. Z. Zhao, V . Kumar, S. Levine, and C. Finn. Learning fine-grained bimanual manipulation with low-cost hardware, 2023","work_id":"940e6de2-97a0-4454-806a-94c0994d1186","year":2023},{"cited_arxiv_id":"2401.02117","doi":"","is_internal_anchor":true,"ref_index":3,"title":"Mobile ALOHA: Learning Bimanual Mobile Manipulation with Low-Cost Whole-Body Teleoperation","work_id":"5f6ff8ef-ed80-4c00-92c2-361c80bf8448","year":2024},{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":4,"title":"R. Bommasani, D. A. Hudson, E. Adeli, R. Altman, S. Arora, S. von Arx, M. S. Bernstein, J. Bohg, A. Bosselut, E. Brunskill, E. Brynjolfsson, S. Buch, D. Card, R. Castellon, N. Chatterji, A. Chen, K. C","work_id":"580d566d-f8f9-4c12-b7f9-35cfad2dbaad","year":2022},{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":5,"title":"A. Brohan, N. Brown, J. Carbajal, Y . Chebotar, X. Chen, K. Choromanski, T. Ding, D. Driess, A. Dubey, C. Finn, P. Florence, C. Fu, M. G. Arenas, K. Gopalakrishnan, K. Han, K. Hausman, A. Herzog, J. H","work_id":"4720ca2c-3947-43dd-8276-6801cea0a7ef","year":2023}],"snapshot_sha256":"0505ae30557932879eb321cad49d6ad1e5108b48d93b96328f83b00a6224ad27"},"source":{"id":"2407.08693","kind":"arxiv","version":3},"verdict":{"created_at":"2026-05-15T08:08:48.921925Z","id":"ba2308ba-b3f6-4bbe-95e5-0f44d7a7f874","model_set":{"reader":"grok-4.3"},"one_line_summary":"Training VLAs to perform embodied chain-of-thought reasoning about plans, sub-tasks, motions, and grounded visual features before acting raises OpenVLA success rates by 28% on challenging generalization tasks without new robot data.","pipeline_version":"pith-pipeline@v0.9.0","pith_extraction_headline":"Embodied chain-of-thought reasoning trains VLAs to output grounded plans and visuals before actions, raising OpenVLA success by 28 percent.","strongest_claim":"ECoT increases the absolute success rate of OpenVLA, the current strongest open-source VLA policy, by 28% across challenging generalization tasks, without any additional robot training data.","weakest_assumption":"The synthetic data generation pipeline produces reasoning traces that are both accurate enough to supervise the model and sufficiently diverse to improve generalization rather than overfitting to the generation heuristics."}},"verdict_id":"ba2308ba-b3f6-4bbe-95e5-0f44d7a7f874"}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:2353a05537249200f8dae62d9d3a129d40d824df3228d32dbced2ed1e9f6e0bf","target":"record","created_at":"2026-05-17T23:38:53Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"attestation_state":"computed","canonical_record":{"metadata":{"abstract_canon_sha256":"5654583cd71b041baba26beae7e5844482c33556f0cc52ddf81e0b4dc4d3ae51","cross_cats_sorted":["cs.LG"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.RO","submitted_at":"2024-07-11T17:31:01Z","title_canon_sha256":"ab4d44c4fd049cf8a78071f32caa91ce066d78cd9a1a1a36d75e8876fdf214b1"},"schema_version":"1.0","source":{"id":"2407.08693","kind":"arxiv","version":3}},"canonical_sha256":"2ad7b65eed6806a29828aad57a07b48081ced4cee339a187b7367e5d52c967da","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"2ad7b65eed6806a29828aad57a07b48081ced4cee339a187b7367e5d52c967da","first_computed_at":"2026-05-17T23:38:53.051031Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-17T23:38:53.051031Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"zCCawEByxAjj+psE6uPCNYxc1z0eAsHh5Og6Bfkxzpw0z2J9ool04nPlfz6BXhaRgI/NTwz01Xjk+n6alWi5CA==","signature_status":"signed_v1","signed_at":"2026-05-17T23:38:53.051526Z","signed_message":"canonical_sha256_bytes"},"source_id":"2407.08693","source_kind":"arxiv","source_version":3}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:2353a05537249200f8dae62d9d3a129d40d824df3228d32dbced2ed1e9f6e0bf","sha256:82fd301cb652a6d54eb6c272900f074729ce007904d2fdc9473b852ffe359aae"],"state_sha256":"018b3dddba1941837ac732e4cf3f3487b5986ff9b881769e99c62330bb72258e"}