{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:3C24JPFCJF3IDQJIWXJ7KVJHJN","short_pith_number":"pith:3C24JPFC","schema_version":"1.0","canonical_sha256":"d8b5c4bca2497681c128b5d3f555274b60d14cf07626e0094f89891ac5bc4e96","source":{"kind":"arxiv","id":"2505.19769","version":2},"attestation_state":"computed","paper":{"title":"TeViR: Text-to-Video Reward with Diffusion Models for Efficient Reinforcement Learning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.RO","authors_text":"Dongbin Zhao, Haoran Li, Haowei Wen, Yuhui Chen, Zhennan Jiang","submitted_at":"2025-05-26T09:52:25Z","abstract_excerpt":"Developing scalable and generalizable reward engineering for reinforcement learning (RL) is crucial for creating general-purpose agents, especially in the challenging domain of robotic manipulation. While recent advances in reward engineering with Vision-Language Models (VLMs) have shown promise, their sparse reward nature significantly limits sample efficiency. This paper introduces TeViR, a novel method that leverages a pre-trained text-to-video diffusion model to generate dense rewards by comparing the predicted image sequence with current observations. Experimental results across 11 comple"},"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":"2505.19769","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.RO","submitted_at":"2025-05-26T09:52:25Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"b2e6ac1312dcff179ed81e2f37acb8252ffe786d652e0e958c4a4219758027c4","abstract_canon_sha256":"7975b8ebe4dc861c6b2ed3b454269c9894d032bbded93c35e116597462d38a35"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:26:04.781562Z","signature_b64":"kRcBaByi9TDGXtMN2BWQQetVPE3bvvZWywo4ILflNck3sl8iIt91GlVt/EXXDwqcWla7ccAZst3ds0/tILxBCQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"d8b5c4bca2497681c128b5d3f555274b60d14cf07626e0094f89891ac5bc4e96","last_reissued_at":"2026-07-05T11:26:04.781056Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:26:04.781056Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"TeViR: Text-to-Video Reward with Diffusion Models for Efficient Reinforcement Learning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.RO","authors_text":"Dongbin Zhao, Haoran Li, Haowei Wen, Yuhui Chen, Zhennan Jiang","submitted_at":"2025-05-26T09:52:25Z","abstract_excerpt":"Developing scalable and generalizable reward engineering for reinforcement learning (RL) is crucial for creating general-purpose agents, especially in the challenging domain of robotic manipulation. While recent advances in reward engineering with Vision-Language Models (VLMs) have shown promise, their sparse reward nature significantly limits sample efficiency. This paper introduces TeViR, a novel method that leverages a pre-trained text-to-video diffusion model to generate dense rewards by comparing the predicted image sequence with current observations. Experimental results across 11 comple"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2505.19769","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/2505.19769/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":"2505.19769","created_at":"2026-07-05T11:26:04.781116+00:00"},{"alias_kind":"arxiv_version","alias_value":"2505.19769v2","created_at":"2026-07-05T11:26:04.781116+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2505.19769","created_at":"2026-07-05T11:26:04.781116+00:00"},{"alias_kind":"pith_short_12","alias_value":"3C24JPFCJF3I","created_at":"2026-07-05T11:26:04.781116+00:00"},{"alias_kind":"pith_short_16","alias_value":"3C24JPFCJF3IDQJI","created_at":"2026-07-05T11:26:04.781116+00:00"},{"alias_kind":"pith_short_8","alias_value":"3C24JPFC","created_at":"2026-07-05T11:26:04.781116+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2602.10503","citing_title":"Towards Long-Lived Robots: Continual Learning VLA Models via Reinforcement Fine-Tuning","ref_index":12,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/3C24JPFCJF3IDQJIWXJ7KVJHJN","json":"https://pith.science/pith/3C24JPFCJF3IDQJIWXJ7KVJHJN.json","graph_json":"https://pith.science/api/pith-number/3C24JPFCJF3IDQJIWXJ7KVJHJN/graph.json","events_json":"https://pith.science/api/pith-number/3C24JPFCJF3IDQJIWXJ7KVJHJN/events.json","paper":"https://pith.science/paper/3C24JPFC"},"agent_actions":{"view_html":"https://pith.science/pith/3C24JPFCJF3IDQJIWXJ7KVJHJN","download_json":"https://pith.science/pith/3C24JPFCJF3IDQJIWXJ7KVJHJN.json","view_paper":"https://pith.science/paper/3C24JPFC","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2505.19769&json=true","fetch_graph":"https://pith.science/api/pith-number/3C24JPFCJF3IDQJIWXJ7KVJHJN/graph.json","fetch_events":"https://pith.science/api/pith-number/3C24JPFCJF3IDQJIWXJ7KVJHJN/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/3C24JPFCJF3IDQJIWXJ7KVJHJN/action/timestamp_anchor","attest_storage":"https://pith.science/pith/3C24JPFCJF3IDQJIWXJ7KVJHJN/action/storage_attestation","attest_author":"https://pith.science/pith/3C24JPFCJF3IDQJIWXJ7KVJHJN/action/author_attestation","sign_citation":"https://pith.science/pith/3C24JPFCJF3IDQJIWXJ7KVJHJN/action/citation_signature","submit_replication":"https://pith.science/pith/3C24JPFCJF3IDQJIWXJ7KVJHJN/action/replication_record"}},"created_at":"2026-07-05T11:26:04.781116+00:00","updated_at":"2026-07-05T11:26:04.781116+00:00"}