{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:Y4FY3MVX7P6B4ZASFDXFJPKCOP","short_pith_number":"pith:Y4FY3MVX","schema_version":"1.0","canonical_sha256":"c70b8db2b7fbfc1e641228ee54bd4273efb12a47e56ebec1b3119ecc8a9eb986","source":{"kind":"arxiv","id":"2411.04549","version":1},"attestation_state":"computed","paper":{"title":"Vision Language Models are In-Context Value Learners","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.LG"],"primary_cat":"cs.RO","authors_text":"Ayzaan Wahid, Chuyuan Fu, Danny Driess, Dhruv Shah, Dinesh Jayaraman, Dorsa Sadigh, Fei Xia, Jacky Liang, Joey Hejna, Jonathan Tompson, Osbert Bastani, Peng Xu, Sean Kirmani, Ted Xiao, Tingnan Zhang, Wenhao Yu, Yecheng Jason Ma, Zhuo Xu","submitted_at":"2024-11-07T09:17:50Z","abstract_excerpt":"Predicting temporal progress from visual trajectories is important for intelligent robots that can learn, adapt, and improve. However, learning such progress estimator, or temporal value function, across different tasks and domains requires both a large amount of diverse data and methods which can scale and generalize. To address these challenges, we present Generative Value Learning (\\GVL), a universal value function estimator that leverages the world knowledge embedded in vision-language models (VLMs) to predict task progress. Naively asking a VLM to predict values for a video sequence perfo"},"verification_status":{"content_addressed":true,"pith_receipt":true,"author_attested":false,"weak_author_claims":0,"strong_author_claims":0,"externally_anchored":false,"storage_verified":false,"citation_signatures":0,"replication_records":0,"graph_snapshot":true,"references_resolved":false,"formal_links_present":false},"canonical_record":{"source":{"id":"2411.04549","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.RO","submitted_at":"2024-11-07T09:17:50Z","cross_cats_sorted":["cs.AI","cs.LG"],"title_canon_sha256":"1636111cd9e778ebf9bc35bbea5cf7dc7e066bc37dda62823bb2916ad0cb477d","abstract_canon_sha256":"5621da42fac54f3d6113833e1f621f1f88592412e4fd15e8dbefbd8eb81224c9"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:32:21.826890Z","signature_b64":"BaV14fRdmAAf4Xwv1PJEB2L5jan6d+P9qbAvnS3n3NH0ecOrlLXWagNSj626w9JJP0/rGuARLc7dQVHn1oxQBg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"c70b8db2b7fbfc1e641228ee54bd4273efb12a47e56ebec1b3119ecc8a9eb986","last_reissued_at":"2026-07-05T09:32:21.826357Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:32:21.826357Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Vision Language Models are In-Context Value Learners","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.LG"],"primary_cat":"cs.RO","authors_text":"Ayzaan Wahid, Chuyuan Fu, Danny Driess, Dhruv Shah, Dinesh Jayaraman, Dorsa Sadigh, Fei Xia, Jacky Liang, Joey Hejna, Jonathan Tompson, Osbert Bastani, Peng Xu, Sean Kirmani, Ted Xiao, Tingnan Zhang, Wenhao Yu, Yecheng Jason Ma, Zhuo Xu","submitted_at":"2024-11-07T09:17:50Z","abstract_excerpt":"Predicting temporal progress from visual trajectories is important for intelligent robots that can learn, adapt, and improve. However, learning such progress estimator, or temporal value function, across different tasks and domains requires both a large amount of diverse data and methods which can scale and generalize. To address these challenges, we present Generative Value Learning (\\GVL), a universal value function estimator that leverages the world knowledge embedded in vision-language models (VLMs) to predict task progress. Naively asking a VLM to predict values for a video sequence perfo"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2411.04549","kind":"arxiv","version":1},"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/2411.04549/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"},"aliases":[{"alias_kind":"arxiv","alias_value":"2411.04549","created_at":"2026-07-05T09:32:21.826438+00:00"},{"alias_kind":"arxiv_version","alias_value":"2411.04549v1","created_at":"2026-07-05T09:32:21.826438+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2411.04549","created_at":"2026-07-05T09:32:21.826438+00:00"},{"alias_kind":"pith_short_12","alias_value":"Y4FY3MVX7P6B","created_at":"2026-07-05T09:32:21.826438+00:00"},{"alias_kind":"pith_short_16","alias_value":"Y4FY3MVX7P6B4ZAS","created_at":"2026-07-05T09:32:21.826438+00:00"},{"alias_kind":"pith_short_8","alias_value":"Y4FY3MVX","created_at":"2026-07-05T09:32:21.826438+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":4,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2607.05391","citing_title":"LLM-as-a-Verifier: A General-Purpose Verification Framework","ref_index":23,"is_internal_anchor":true},{"citing_arxiv_id":"2606.21572","citing_title":"Robot Critics that Sweat the Small Stuff","ref_index":9,"is_internal_anchor":false},{"citing_arxiv_id":"2606.13675","citing_title":"Improving Robotic Generalist Policies via Flow Reversal Steering","ref_index":40,"is_internal_anchor":false},{"citing_arxiv_id":"2605.08774","citing_title":"ProcVLM: Learning Procedure-Grounded Progress Rewards for Robotic Manipulation","ref_index":45,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/Y4FY3MVX7P6B4ZASFDXFJPKCOP","json":"https://pith.science/pith/Y4FY3MVX7P6B4ZASFDXFJPKCOP.json","graph_json":"https://pith.science/api/pith-number/Y4FY3MVX7P6B4ZASFDXFJPKCOP/graph.json","events_json":"https://pith.science/api/pith-number/Y4FY3MVX7P6B4ZASFDXFJPKCOP/events.json","paper":"https://pith.science/paper/Y4FY3MVX"},"agent_actions":{"view_html":"https://pith.science/pith/Y4FY3MVX7P6B4ZASFDXFJPKCOP","download_json":"https://pith.science/pith/Y4FY3MVX7P6B4ZASFDXFJPKCOP.json","view_paper":"https://pith.science/paper/Y4FY3MVX","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2411.04549&json=true","fetch_graph":"https://pith.science/api/pith-number/Y4FY3MVX7P6B4ZASFDXFJPKCOP/graph.json","fetch_events":"https://pith.science/api/pith-number/Y4FY3MVX7P6B4ZASFDXFJPKCOP/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/Y4FY3MVX7P6B4ZASFDXFJPKCOP/action/timestamp_anchor","attest_storage":"https://pith.science/pith/Y4FY3MVX7P6B4ZASFDXFJPKCOP/action/storage_attestation","attest_author":"https://pith.science/pith/Y4FY3MVX7P6B4ZASFDXFJPKCOP/action/author_attestation","sign_citation":"https://pith.science/pith/Y4FY3MVX7P6B4ZASFDXFJPKCOP/action/citation_signature","submit_replication":"https://pith.science/pith/Y4FY3MVX7P6B4ZASFDXFJPKCOP/action/replication_record"}},"created_at":"2026-07-05T09:32:21.826438+00:00","updated_at":"2026-07-05T09:32:21.826438+00:00"}