{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2023:OLU5FDOJEC3JOCAOJTGUKGD2AG","short_pith_number":"pith:OLU5FDOJ","canonical_record":{"source":{"id":"2305.14343","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2023-05-23T17:59:33Z","cross_cats_sorted":["cs.AI","cs.CV"],"title_canon_sha256":"49b9fa6a17d4e4ffe233168991c56648d457c956af293d62ebe92cbc0bee4593","abstract_canon_sha256":"93dcd3e5ad0a943bc236e2b89e3b968061c01601bad74841a7fbbdadf8d8028e"},"schema_version":"1.0"},"canonical_sha256":"72e9d28dc920b697080e4ccd45187a01a5dd0745a126c092fe6df7b95e115459","source":{"kind":"arxiv","id":"2305.14343","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2305.14343","created_at":"2026-07-05T06:15:32Z"},{"alias_kind":"arxiv_version","alias_value":"2305.14343v2","created_at":"2026-07-05T06:15:32Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2305.14343","created_at":"2026-07-05T06:15:32Z"},{"alias_kind":"pith_short_12","alias_value":"OLU5FDOJEC3J","created_at":"2026-07-05T06:15:32Z"},{"alias_kind":"pith_short_16","alias_value":"OLU5FDOJEC3JOCAO","created_at":"2026-07-05T06:15:32Z"},{"alias_kind":"pith_short_8","alias_value":"OLU5FDOJ","created_at":"2026-07-05T06:15:32Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2023:OLU5FDOJEC3JOCAOJTGUKGD2AG","target":"record","payload":{"canonical_record":{"source":{"id":"2305.14343","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2023-05-23T17:59:33Z","cross_cats_sorted":["cs.AI","cs.CV"],"title_canon_sha256":"49b9fa6a17d4e4ffe233168991c56648d457c956af293d62ebe92cbc0bee4593","abstract_canon_sha256":"93dcd3e5ad0a943bc236e2b89e3b968061c01601bad74841a7fbbdadf8d8028e"},"schema_version":"1.0"},"canonical_sha256":"72e9d28dc920b697080e4ccd45187a01a5dd0745a126c092fe6df7b95e115459","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T06:15:32.454286Z","signature_b64":"g66Cf8z9CV4vuxDySAcgq91vf667FtllQe9mjof5UPUOdMvjLlMtRvePe8GtVytq99jIa5KU44jguvA2ZVTgBA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"72e9d28dc920b697080e4ccd45187a01a5dd0745a126c092fe6df7b95e115459","last_reissued_at":"2026-07-05T06:15:32.453779Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T06:15:32.453779Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2305.14343","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":"2026-07-05T06:15:32Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"2lQ7v6hOjtnE/9n6y+b9W6FqfshYqxYoYKU7GRHFJyjbKDrKooIk3QIzjiQ26UxgnDgOVIVN7keXOjvwa8m2DQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-01T01:20:01.710273Z"},"content_sha256":"6422da00e30c1cfb3684d5396f407f185b73a9aeced170dc4d11765b40a8d17f","schema_version":"1.0","event_id":"sha256:6422da00e30c1cfb3684d5396f407f185b73a9aeced170dc4d11765b40a8d17f"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2023:OLU5FDOJEC3JOCAOJTGUKGD2AG","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Video Prediction Models as Rewards for Reinforcement Learning","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.CV"],"primary_cat":"cs.LG","authors_text":"Ademi Adeniji, Ajay Jain, Alejandro Escontrela, Danijar Hafner, Ken Goldberg, Pieter Abbeel, Wilson Yan, Xue Bin Peng, Youngwoon Lee","submitted_at":"2023-05-23T17:59:33Z","abstract_excerpt":"Specifying reward signals that allow agents to learn complex behaviors is a long-standing challenge in reinforcement learning. A promising approach is to extract preferences for behaviors from unlabeled videos, which are widely available on the internet. We present Video Prediction Rewards (VIPER), an algorithm that leverages pretrained video prediction models as action-free reward signals for reinforcement learning. Specifically, we first train an autoregressive transformer on expert videos and then use the video prediction likelihoods as reward signals for a reinforcement learning agent. VIP"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2305.14343","kind":"arxiv","version":2},"verdict":{"id":null,"model_set":{},"created_at":null,"strongest_claim":"","one_line_summary":"","pipeline_version":null,"weakest_assumption":"","pith_extraction_headline":""},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2305.14343/integrity.json","findings":[],"available":true,"detectors_run":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938"},"references":{"count":0,"sample":[],"resolved_work":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","internal_anchors":0},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"author_claims":{"count":0,"strong_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"builder_version":"pith-number-builder-2026-05-17-v1"},"verdict_id":null},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T06:15:32Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"0dhma1u0i0wevuY0qRCmqrbWRLEmVBuZvITTXlW/WkSBYKaVcTiK9c5OE1xxivFp+in3OpQiV8aFQ2BIzA7JAA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-01T01:20:01.714697Z"},"content_sha256":"5ea5f3a2ad35b1728167f62c5c699a807a50049d7d2e8518ba187cfd59c8aa0b","schema_version":"1.0","event_id":"sha256:5ea5f3a2ad35b1728167f62c5c699a807a50049d7d2e8518ba187cfd59c8aa0b"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/OLU5FDOJEC3JOCAOJTGUKGD2AG/bundle.json","state_url":"https://pith.science/pith/OLU5FDOJEC3JOCAOJTGUKGD2AG/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/OLU5FDOJEC3JOCAOJTGUKGD2AG/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":"2026-08-01T01:20:01Z","links":{"resolver":"https://pith.science/pith/OLU5FDOJEC3JOCAOJTGUKGD2AG","bundle":"https://pith.science/pith/OLU5FDOJEC3JOCAOJTGUKGD2AG/bundle.json","state":"https://pith.science/pith/OLU5FDOJEC3JOCAOJTGUKGD2AG/state.json","well_known_bundle":"https://pith.science/.well-known/pith/OLU5FDOJEC3JOCAOJTGUKGD2AG/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2023:OLU5FDOJEC3JOCAOJTGUKGD2AG","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":"93dcd3e5ad0a943bc236e2b89e3b968061c01601bad74841a7fbbdadf8d8028e","cross_cats_sorted":["cs.AI","cs.CV"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2023-05-23T17:59:33Z","title_canon_sha256":"49b9fa6a17d4e4ffe233168991c56648d457c956af293d62ebe92cbc0bee4593"},"schema_version":"1.0","source":{"id":"2305.14343","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2305.14343","created_at":"2026-07-05T06:15:32Z"},{"alias_kind":"arxiv_version","alias_value":"2305.14343v2","created_at":"2026-07-05T06:15:32Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2305.14343","created_at":"2026-07-05T06:15:32Z"},{"alias_kind":"pith_short_12","alias_value":"OLU5FDOJEC3J","created_at":"2026-07-05T06:15:32Z"},{"alias_kind":"pith_short_16","alias_value":"OLU5FDOJEC3JOCAO","created_at":"2026-07-05T06:15:32Z"},{"alias_kind":"pith_short_8","alias_value":"OLU5FDOJ","created_at":"2026-07-05T06:15:32Z"}],"graph_snapshots":[{"event_id":"sha256:5ea5f3a2ad35b1728167f62c5c699a807a50049d7d2e8518ba187cfd59c8aa0b","target":"graph","created_at":"2026-07-05T06:15:32Z","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":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"integrity":{"available":true,"clean":true,"detectors_run":[],"endpoint":"/pith/2305.14343/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Specifying reward signals that allow agents to learn complex behaviors is a long-standing challenge in reinforcement learning. A promising approach is to extract preferences for behaviors from unlabeled videos, which are widely available on the internet. We present Video Prediction Rewards (VIPER), an algorithm that leverages pretrained video prediction models as action-free reward signals for reinforcement learning. Specifically, we first train an autoregressive transformer on expert videos and then use the video prediction likelihoods as reward signals for a reinforcement learning agent. VIP","authors_text":"Ademi Adeniji, Ajay Jain, Alejandro Escontrela, Danijar Hafner, Ken Goldberg, Pieter Abbeel, Wilson Yan, Xue Bin Peng, Youngwoon Lee","cross_cats":["cs.AI","cs.CV"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2023-05-23T17:59:33Z","title":"Video Prediction Models as Rewards for Reinforcement Learning"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2305.14343","kind":"arxiv","version":2},"verdict":{"created_at":null,"id":null,"model_set":{},"one_line_summary":"","pipeline_version":null,"pith_extraction_headline":"","strongest_claim":"","weakest_assumption":""}},"verdict_id":null}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:6422da00e30c1cfb3684d5396f407f185b73a9aeced170dc4d11765b40a8d17f","target":"record","created_at":"2026-07-05T06:15:32Z","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":"93dcd3e5ad0a943bc236e2b89e3b968061c01601bad74841a7fbbdadf8d8028e","cross_cats_sorted":["cs.AI","cs.CV"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2023-05-23T17:59:33Z","title_canon_sha256":"49b9fa6a17d4e4ffe233168991c56648d457c956af293d62ebe92cbc0bee4593"},"schema_version":"1.0","source":{"id":"2305.14343","kind":"arxiv","version":2}},"canonical_sha256":"72e9d28dc920b697080e4ccd45187a01a5dd0745a126c092fe6df7b95e115459","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"72e9d28dc920b697080e4ccd45187a01a5dd0745a126c092fe6df7b95e115459","first_computed_at":"2026-07-05T06:15:32.453779Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T06:15:32.453779Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"g66Cf8z9CV4vuxDySAcgq91vf667FtllQe9mjof5UPUOdMvjLlMtRvePe8GtVytq99jIa5KU44jguvA2ZVTgBA==","signature_status":"signed_v1","signed_at":"2026-07-05T06:15:32.454286Z","signed_message":"canonical_sha256_bytes"},"source_id":"2305.14343","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:6422da00e30c1cfb3684d5396f407f185b73a9aeced170dc4d11765b40a8d17f","sha256:5ea5f3a2ad35b1728167f62c5c699a807a50049d7d2e8518ba187cfd59c8aa0b"],"state_sha256":"fe66b31be350d4db8fb7cf4065f83041e1c4bc728f407adaa8005e90ba0b69af"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"1KzUYtUZ49SRd44H6T4NY321ltNWg4qPkx1ABN0/n1YqPyCn70ETuAnEvPXjEdNdjZ85Iz1xzlN5wDHbJV8iDg==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-01T01:20:01.719000Z","bundle_sha256":"c638f1cff98a53fcf9e79f7d7fa4c0c766ef1239aef854e551b44047477c790a"}}