{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2024:TOPMBZWI6RZ5XX2DT5CY6BIJNU","short_pith_number":"pith:TOPMBZWI","canonical_record":{"source":{"id":"2410.11758","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.RO","submitted_at":"2024-10-15T16:28:09Z","cross_cats_sorted":["cs.CL","cs.CV","cs.LG"],"title_canon_sha256":"697b959e671bb5612b90d4d518b6f82f95585f0178d02046d14fbe7347b291a6","abstract_canon_sha256":"c6fad7a3cbbc7d753ab0f1d39139b2cafc029ca7da297eb88a9da486880d7cbb"},"schema_version":"1.0"},"canonical_sha256":"9b9ec0e6c8f473dbdf439f458f05096d27563615659ffcb6b3014e1b29554d09","source":{"kind":"arxiv","id":"2410.11758","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2410.11758","created_at":"2026-05-18T02:36:37Z"},{"alias_kind":"arxiv_version","alias_value":"2410.11758v2","created_at":"2026-05-18T02:36:37Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2410.11758","created_at":"2026-05-18T02:36:37Z"},{"alias_kind":"pith_short_12","alias_value":"TOPMBZWI6RZ5","created_at":"2026-05-18T12:33:37Z"},{"alias_kind":"pith_short_16","alias_value":"TOPMBZWI6RZ5XX2D","created_at":"2026-05-18T12:33:37Z"},{"alias_kind":"pith_short_8","alias_value":"TOPMBZWI","created_at":"2026-05-18T12:33:37Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2024:TOPMBZWI6RZ5XX2DT5CY6BIJNU","target":"record","payload":{"canonical_record":{"source":{"id":"2410.11758","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.RO","submitted_at":"2024-10-15T16:28:09Z","cross_cats_sorted":["cs.CL","cs.CV","cs.LG"],"title_canon_sha256":"697b959e671bb5612b90d4d518b6f82f95585f0178d02046d14fbe7347b291a6","abstract_canon_sha256":"c6fad7a3cbbc7d753ab0f1d39139b2cafc029ca7da297eb88a9da486880d7cbb"},"schema_version":"1.0"},"canonical_sha256":"9b9ec0e6c8f473dbdf439f458f05096d27563615659ffcb6b3014e1b29554d09","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-18T02:36:37.771795Z","signature_b64":"Th7JDnaJ5voq/B+SAjifz65TIOvfk0nczGb/uap3ui6Yf/XZHjrdrZyQtzTWZQ4u9URGt+hIjEMKnvJSxyqvAg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"9b9ec0e6c8f473dbdf439f458f05096d27563615659ffcb6b3014e1b29554d09","last_reissued_at":"2026-05-18T02:36:37.771191Z","signature_status":"signed_v1","first_computed_at":"2026-05-18T02:36:37.771191Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2410.11758","source_version":2,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-05-18T02:36:37Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"gZD/OacpX+kNds9/eer5i51FYH1XfIocbSFD+lAHgEuFN9UENzydgAouNbxMnWH0+oVZpM/JukH05Q+EK/a0Bg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-06-10T23:41:09.006669Z"},"content_sha256":"01a2f760a175f8984dcc3089030a3d86c34a1ed85047caa0206c10cd28523c2b","schema_version":"1.0","event_id":"sha256:01a2f760a175f8984dcc3089030a3d86c34a1ed85047caa0206c10cd28523c2b"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2024:TOPMBZWI6RZ5XX2DT5CY6BIJNU","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Latent Action Pretraining from Videos","license":"http://creativecommons.org/licenses/by/4.0/","headline":"LAPA learns discrete latent actions from unlabeled videos with VQ-VAE, pretrains a VLA model to predict them, and finetunes on small robot datasets to outperform both video-only baselines and labeled SOTA VLA models on language-conditioned manipulation tasks.","cross_cats":["cs.CL","cs.CV","cs.LG"],"primary_cat":"cs.RO","authors_text":"Ajay Mandlekar, Baolin Peng, Bill Yuchen Lin, Byeongguk Jeon, Dieter Fox, Jianfeng Gao, Jianwei Yang, Joel Jang, Kimin Lee, Lars Liden, Luke Zettlemoyer, Minjoon Seo, Reuben Tan, Sejune Joo, Seonghyeon Ye, Yu-Wei Chao","submitted_at":"2024-10-15T16:28:09Z","abstract_excerpt":"We introduce Latent Action Pretraining for general Action models (LAPA), an unsupervised method for pretraining Vision-Language-Action (VLA) models without ground-truth robot action labels. Existing Vision-Language-Action models require action labels typically collected by human teleoperators during pretraining, which significantly limits possible data sources and scale. In this work, we propose a method to learn from internet-scale videos that do not have robot action labels. We first train an action quantization model leveraging VQ-VAE-based objective to learn discrete latent actions between"},"claims":{"count":3,"items":[{"kind":"strongest_claim","text":"Experimental results demonstrate that our method significantly outperforms existing techniques that train robot manipulation policies from large-scale videos. Furthermore, it outperforms the state-of-the-art VLA model trained with robotic action labels on real-world manipulation tasks that require language conditioning, generalization to unseen objects, and semantic generalization to unseen instructions.","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","text":"That discrete latent actions extracted from human manipulation videos contain sufficient transferable information to map effectively to robot actions during finetuning and yield better generalization than direct supervised training on labeled robot data.","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","text":"LAPA learns discrete latent actions from unlabeled videos with VQ-VAE, pretrains a VLA model to predict them, and finetunes on small robot datasets to outperform both video-only baselines and labeled SOTA VLA models on language-conditioned manipulation tasks.","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"}],"snapshot_sha256":"1dd1973c2177d350f72c677c66595154b23809b747fb41d0ae158dbb0bcf80d5"},"source":{"id":"2410.11758","kind":"arxiv","version":2},"verdict":{"id":"f5188e97-42fa-4359-bebc-1948939438a6","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-14T02:23:37.058623Z","strongest_claim":"Experimental results demonstrate that our method significantly outperforms existing techniques that train robot manipulation policies from large-scale videos. Furthermore, it outperforms the state-of-the-art VLA model trained with robotic action labels on real-world manipulation tasks that require language conditioning, generalization to unseen objects, and semantic generalization to unseen instructions.","one_line_summary":"LAPA learns discrete latent actions from unlabeled videos with VQ-VAE, pretrains a VLA model to predict them, and finetunes on small robot datasets to outperform both video-only baselines and labeled SOTA VLA models on language-conditioned manipulation tasks.","pipeline_version":"pith-pipeline@v0.9.0","weakest_assumption":"That discrete latent actions extracted from human manipulation videos contain sufficient transferable information to map effectively to robot actions during finetuning and yield better generalization than direct supervised training on labeled robot data.","pith_extraction_headline":""},"references":{"count":0,"sample":[],"resolved_work":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","internal_anchors":0},"formal_canon":{"evidence_count":2,"snapshot_sha256":"c5bc0f3d18b0aa144e8d64d0cbcc47c58090d38db979e35ae20bd97ea25f93f6"},"author_claims":{"count":0,"strong_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"builder_version":"pith-number-builder-2026-05-17-v1"},"verdict_id":"f5188e97-42fa-4359-bebc-1948939438a6"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-05-18T02:36:37Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"xc2NtjYmNM1Q7gHrBldrGu55u3QSEwnR9OKxZ/1cvWrEYkerrOPT9roIz0GVzm7dJFFYygKv9oLjQSoTsUdHAg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-06-10T23:41:09.007550Z"},"content_sha256":"fd07156649d16126431d105539d73d53ba65c02e9c03d8372a6f5584492bbd28","schema_version":"1.0","event_id":"sha256:fd07156649d16126431d105539d73d53ba65c02e9c03d8372a6f5584492bbd28"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/TOPMBZWI6RZ5XX2DT5CY6BIJNU/bundle.json","state_url":"https://pith.science/pith/TOPMBZWI6RZ5XX2DT5CY6BIJNU/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/TOPMBZWI6RZ5XX2DT5CY6BIJNU/bundle.json","status":"primary"}],"public_keys":[{"key_id":"pith-v1-2026-05","algorithm":"ed25519","format":"raw","public_key_b64":"stVStoiQhXFxp4s2pdzPNoqVNBMojDU/fJ2db5S3CbM=","public_key_hex":"b2d552b68890857171a78b36a5dccf368a953413288c353f7c9d9d6f94b709b3","fingerprint_sha256_b32_first128bits":"RVFV5Z2OI2J3ZUO7ERDEBCYNKS","fingerprint_sha256_hex":"8d4b5ee74e4693bcd1df2446408b0d54","rotates_at":null,"url":"https://pith.science/pith-signing-key.json","notes":"Pith uses this Ed25519 key to sign canonical record SHA-256 digests. Verify with: ed25519_verify(public_key, message=canonical_sha256_bytes, signature=base64decode(signature_b64))."}],"merge_version":"pith-open-graph-merge-v1","built_at":"2026-06-10T23:41:09Z","links":{"resolver":"https://pith.science/pith/TOPMBZWI6RZ5XX2DT5CY6BIJNU","bundle":"https://pith.science/pith/TOPMBZWI6RZ5XX2DT5CY6BIJNU/bundle.json","state":"https://pith.science/pith/TOPMBZWI6RZ5XX2DT5CY6BIJNU/state.json","well_known_bundle":"https://pith.science/.well-known/pith/TOPMBZWI6RZ5XX2DT5CY6BIJNU/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:TOPMBZWI6RZ5XX2DT5CY6BIJNU","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":"c6fad7a3cbbc7d753ab0f1d39139b2cafc029ca7da297eb88a9da486880d7cbb","cross_cats_sorted":["cs.CL","cs.CV","cs.LG"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.RO","submitted_at":"2024-10-15T16:28:09Z","title_canon_sha256":"697b959e671bb5612b90d4d518b6f82f95585f0178d02046d14fbe7347b291a6"},"schema_version":"1.0","source":{"id":"2410.11758","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2410.11758","created_at":"2026-05-18T02:36:37Z"},{"alias_kind":"arxiv_version","alias_value":"2410.11758v2","created_at":"2026-05-18T02:36:37Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2410.11758","created_at":"2026-05-18T02:36:37Z"},{"alias_kind":"pith_short_12","alias_value":"TOPMBZWI6RZ5","created_at":"2026-05-18T12:33:37Z"},{"alias_kind":"pith_short_16","alias_value":"TOPMBZWI6RZ5XX2D","created_at":"2026-05-18T12:33:37Z"},{"alias_kind":"pith_short_8","alias_value":"TOPMBZWI","created_at":"2026-05-18T12:33:37Z"}],"graph_snapshots":[{"event_id":"sha256:fd07156649d16126431d105539d73d53ba65c02e9c03d8372a6f5584492bbd28","target":"graph","created_at":"2026-05-18T02:36:37Z","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":3,"items":[{"attestation":"unclaimed","claim_id":"C1","kind":"strongest_claim","source":"verdict.strongest_claim","status":"machine_extracted","text":"Experimental results demonstrate that our method significantly outperforms existing techniques that train robot manipulation policies from large-scale videos. Furthermore, it outperforms the state-of-the-art VLA model trained with robotic action labels on real-world manipulation tasks that require language conditioning, generalization to unseen objects, and semantic generalization to unseen instructions."},{"attestation":"unclaimed","claim_id":"C2","kind":"weakest_assumption","source":"verdict.weakest_assumption","status":"machine_extracted","text":"That discrete latent actions extracted from human manipulation videos contain sufficient transferable information to map effectively to robot actions during finetuning and yield better generalization than direct supervised training on labeled robot data."},{"attestation":"unclaimed","claim_id":"C3","kind":"one_line_summary","source":"verdict.one_line_summary","status":"machine_extracted","text":"LAPA learns discrete latent actions from unlabeled videos with VQ-VAE, pretrains a VLA model to predict them, and finetunes on small robot datasets to outperform both video-only baselines and labeled SOTA VLA models on language-conditioned manipulation tasks."}],"snapshot_sha256":"1dd1973c2177d350f72c677c66595154b23809b747fb41d0ae158dbb0bcf80d5"},"formal_canon":{"evidence_count":2,"snapshot_sha256":"c5bc0f3d18b0aa144e8d64d0cbcc47c58090d38db979e35ae20bd97ea25f93f6"},"paper":{"abstract_excerpt":"We introduce Latent Action Pretraining for general Action models (LAPA), an unsupervised method for pretraining Vision-Language-Action (VLA) models without ground-truth robot action labels. Existing Vision-Language-Action models require action labels typically collected by human teleoperators during pretraining, which significantly limits possible data sources and scale. In this work, we propose a method to learn from internet-scale videos that do not have robot action labels. We first train an action quantization model leveraging VQ-VAE-based objective to learn discrete latent actions between","authors_text":"Ajay Mandlekar, Baolin Peng, Bill Yuchen Lin, Byeongguk Jeon, Dieter Fox, Jianfeng Gao, Jianwei Yang, Joel Jang, Kimin Lee, Lars Liden, Luke Zettlemoyer, Minjoon Seo, Reuben Tan, Sejune Joo, Seonghyeon Ye, Yu-Wei Chao","cross_cats":["cs.CL","cs.CV","cs.LG"],"headline":"LAPA learns discrete latent actions from unlabeled videos with VQ-VAE, pretrains a VLA model to predict them, and finetunes on small robot datasets to outperform both video-only baselines and labeled SOTA VLA models on language-conditioned manipulation tasks.","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.RO","submitted_at":"2024-10-15T16:28:09Z","title":"Latent Action Pretraining from Videos"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2410.11758","kind":"arxiv","version":2},"verdict":{"created_at":"2026-05-14T02:23:37.058623Z","id":"f5188e97-42fa-4359-bebc-1948939438a6","model_set":{"reader":"grok-4.3"},"one_line_summary":"LAPA learns discrete latent actions from unlabeled videos with VQ-VAE, pretrains a VLA model to predict them, and finetunes on small robot datasets to outperform both video-only baselines and labeled SOTA VLA models on language-conditioned manipulation tasks.","pipeline_version":"pith-pipeline@v0.9.0","pith_extraction_headline":"","strongest_claim":"Experimental results demonstrate that our method significantly outperforms existing techniques that train robot manipulation policies from large-scale videos. Furthermore, it outperforms the state-of-the-art VLA model trained with robotic action labels on real-world manipulation tasks that require language conditioning, generalization to unseen objects, and semantic generalization to unseen instructions.","weakest_assumption":"That discrete latent actions extracted from human manipulation videos contain sufficient transferable information to map effectively to robot actions during finetuning and yield better generalization than direct supervised training on labeled robot data."}},"verdict_id":"f5188e97-42fa-4359-bebc-1948939438a6"}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:01a2f760a175f8984dcc3089030a3d86c34a1ed85047caa0206c10cd28523c2b","target":"record","created_at":"2026-05-18T02:36:37Z","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":"c6fad7a3cbbc7d753ab0f1d39139b2cafc029ca7da297eb88a9da486880d7cbb","cross_cats_sorted":["cs.CL","cs.CV","cs.LG"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.RO","submitted_at":"2024-10-15T16:28:09Z","title_canon_sha256":"697b959e671bb5612b90d4d518b6f82f95585f0178d02046d14fbe7347b291a6"},"schema_version":"1.0","source":{"id":"2410.11758","kind":"arxiv","version":2}},"canonical_sha256":"9b9ec0e6c8f473dbdf439f458f05096d27563615659ffcb6b3014e1b29554d09","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"9b9ec0e6c8f473dbdf439f458f05096d27563615659ffcb6b3014e1b29554d09","first_computed_at":"2026-05-18T02:36:37.771191Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-18T02:36:37.771191Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"Th7JDnaJ5voq/B+SAjifz65TIOvfk0nczGb/uap3ui6Yf/XZHjrdrZyQtzTWZQ4u9URGt+hIjEMKnvJSxyqvAg==","signature_status":"signed_v1","signed_at":"2026-05-18T02:36:37.771795Z","signed_message":"canonical_sha256_bytes"},"source_id":"2410.11758","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:01a2f760a175f8984dcc3089030a3d86c34a1ed85047caa0206c10cd28523c2b","sha256:fd07156649d16126431d105539d73d53ba65c02e9c03d8372a6f5584492bbd28"],"state_sha256":"a1f1523746d9d056d6db4916b39ec42d9689e340e2f49b2e6c07514fe8107d98"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"vw0grywCqzPEGmjOBcPzmTN7fNKTl5oT8+L8PUdnAxoMexHJLivGoXlGMlfSP4LoAC6H6ahOe4SlqyYwSSgRDQ==","signed_message":"bundle_sha256_bytes","signed_at":"2026-06-10T23:41:09.012200Z","bundle_sha256":"160f942d1437f294f4e51a885c80caa0497923c8687700e42eb92c54755abb38"}}