{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2024:IIBU7NEKIIHSLK4NTGVYOGLFOE","short_pith_number":"pith:IIBU7NEK","canonical_record":{"source":{"id":"2410.04734","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","primary_cat":"cs.LG","submitted_at":"2024-10-07T04:00:22Z","cross_cats_sorted":["cs.CL","cs.CV"],"title_canon_sha256":"b8d6ebb4d630dd96320c2b223d7089816d271f2c2167363f5568bfeb1b67b43c","abstract_canon_sha256":"28eea45c2f9f356381b0d2e07bccb76546a372b50172e0819a4e4fad4aa3023a"},"schema_version":"1.0"},"canonical_sha256":"42034fb48a420f25ab8d99ab871965711eee473deba295cced71455a28c84056","source":{"kind":"arxiv","id":"2410.04734","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2410.04734","created_at":"2026-07-05T10:19:29Z"},{"alias_kind":"arxiv_version","alias_value":"2410.04734v2","created_at":"2026-07-05T10:19:29Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2410.04734","created_at":"2026-07-05T10:19:29Z"},{"alias_kind":"pith_short_12","alias_value":"IIBU7NEKIIHS","created_at":"2026-07-05T10:19:29Z"},{"alias_kind":"pith_short_16","alias_value":"IIBU7NEKIIHSLK4N","created_at":"2026-07-05T10:19:29Z"},{"alias_kind":"pith_short_8","alias_value":"IIBU7NEK","created_at":"2026-07-05T10:19:29Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2024:IIBU7NEKIIHSLK4NTGVYOGLFOE","target":"record","payload":{"canonical_record":{"source":{"id":"2410.04734","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","primary_cat":"cs.LG","submitted_at":"2024-10-07T04:00:22Z","cross_cats_sorted":["cs.CL","cs.CV"],"title_canon_sha256":"b8d6ebb4d630dd96320c2b223d7089816d271f2c2167363f5568bfeb1b67b43c","abstract_canon_sha256":"28eea45c2f9f356381b0d2e07bccb76546a372b50172e0819a4e4fad4aa3023a"},"schema_version":"1.0"},"canonical_sha256":"42034fb48a420f25ab8d99ab871965711eee473deba295cced71455a28c84056","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:19:29.234570Z","signature_b64":"kh7WA4a1Vl267646aVg5QIryTvvTTSEOG1Xc/HhOxKA4kIpHA2wx395IUYA2bDMFg4MqeLIsfMrieN/ecmUaCw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"42034fb48a420f25ab8d99ab871965711eee473deba295cced71455a28c84056","last_reissued_at":"2026-07-05T10:19:29.234068Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:19:29.234068Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2410.04734","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":"2026-07-05T10:19:29Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"lnU5GdKPoGKYRJF/5nbyYp9uWQB2slkU4fq6/X06QZa+wde2InlPdBz8PUovmbImgCHEGkhZvMU0Gp8G9KhkCA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-04T20:51:16.957278Z"},"content_sha256":"e89208672599a0b799c60d8b3895d25dcf4862cba49a59a6bef086b6fe9d8293","schema_version":"1.0","event_id":"sha256:e89208672599a0b799c60d8b3895d25dcf4862cba49a59a6bef086b6fe9d8293"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2024:IIBU7NEKIIHSLK4NTGVYOGLFOE","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"TLDR: Token-Level Detective Reward Model for Large Vision Language Models","license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","headline":"","cross_cats":["cs.CL","cs.CV"],"primary_cat":"cs.LG","authors_text":"Deqing Fu, Guan Pang, Lawrence Chen, Pengchuan Zhang, Robin Jia, Rui Wang, Tong Xiao, Wang Zhu","submitted_at":"2024-10-07T04:00:22Z","abstract_excerpt":"Although reward models have been successful in improving multimodal large language models, the reward models themselves remain brutal and contain minimal information. Notably, existing reward models only mimic human annotations by assigning only one binary feedback to any text, no matter how long the text is. In the realm of multimodal language models, where models are required to process both images and texts, a naive reward model may learn implicit biases toward texts and become less grounded in images. In this paper, we propose a $\\textbf{T}$oken-$\\textbf{L}$evel $\\textbf{D}$etective $\\text"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2410.04734","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/2410.04734/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":"2026-07-05T10:19:29Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"ky48wfnBPPAtJWvRjb+IrQoegl3gcvfbxaKb+d62g73PTNA//KS3hM1ZjgSsPLs5i1NG+oaZnw8ai4y1xdwrBA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-04T20:51:16.957874Z"},"content_sha256":"0e605fb15bc23723e14c06b67827091924fb4be8a5c8edfbaa9e008d3da5af68","schema_version":"1.0","event_id":"sha256:0e605fb15bc23723e14c06b67827091924fb4be8a5c8edfbaa9e008d3da5af68"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/IIBU7NEKIIHSLK4NTGVYOGLFOE/bundle.json","state_url":"https://pith.science/pith/IIBU7NEKIIHSLK4NTGVYOGLFOE/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/IIBU7NEKIIHSLK4NTGVYOGLFOE/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":"2026-08-04T20:51:16Z","links":{"resolver":"https://pith.science/pith/IIBU7NEKIIHSLK4NTGVYOGLFOE","bundle":"https://pith.science/pith/IIBU7NEKIIHSLK4NTGVYOGLFOE/bundle.json","state":"https://pith.science/pith/IIBU7NEKIIHSLK4NTGVYOGLFOE/state.json","well_known_bundle":"https://pith.science/.well-known/pith/IIBU7NEKIIHSLK4NTGVYOGLFOE/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:IIBU7NEKIIHSLK4NTGVYOGLFOE","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":"28eea45c2f9f356381b0d2e07bccb76546a372b50172e0819a4e4fad4aa3023a","cross_cats_sorted":["cs.CL","cs.CV"],"license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","primary_cat":"cs.LG","submitted_at":"2024-10-07T04:00:22Z","title_canon_sha256":"b8d6ebb4d630dd96320c2b223d7089816d271f2c2167363f5568bfeb1b67b43c"},"schema_version":"1.0","source":{"id":"2410.04734","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2410.04734","created_at":"2026-07-05T10:19:29Z"},{"alias_kind":"arxiv_version","alias_value":"2410.04734v2","created_at":"2026-07-05T10:19:29Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2410.04734","created_at":"2026-07-05T10:19:29Z"},{"alias_kind":"pith_short_12","alias_value":"IIBU7NEKIIHS","created_at":"2026-07-05T10:19:29Z"},{"alias_kind":"pith_short_16","alias_value":"IIBU7NEKIIHSLK4N","created_at":"2026-07-05T10:19:29Z"},{"alias_kind":"pith_short_8","alias_value":"IIBU7NEK","created_at":"2026-07-05T10:19:29Z"}],"graph_snapshots":[{"event_id":"sha256:0e605fb15bc23723e14c06b67827091924fb4be8a5c8edfbaa9e008d3da5af68","target":"graph","created_at":"2026-07-05T10:19:29Z","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/2410.04734/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Although reward models have been successful in improving multimodal large language models, the reward models themselves remain brutal and contain minimal information. Notably, existing reward models only mimic human annotations by assigning only one binary feedback to any text, no matter how long the text is. In the realm of multimodal language models, where models are required to process both images and texts, a naive reward model may learn implicit biases toward texts and become less grounded in images. In this paper, we propose a $\\textbf{T}$oken-$\\textbf{L}$evel $\\textbf{D}$etective $\\text","authors_text":"Deqing Fu, Guan Pang, Lawrence Chen, Pengchuan Zhang, Robin Jia, Rui Wang, Tong Xiao, Wang Zhu","cross_cats":["cs.CL","cs.CV"],"headline":"","license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","primary_cat":"cs.LG","submitted_at":"2024-10-07T04:00:22Z","title":"TLDR: Token-Level Detective Reward Model for Large Vision Language Models"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2410.04734","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:e89208672599a0b799c60d8b3895d25dcf4862cba49a59a6bef086b6fe9d8293","target":"record","created_at":"2026-07-05T10:19:29Z","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":"28eea45c2f9f356381b0d2e07bccb76546a372b50172e0819a4e4fad4aa3023a","cross_cats_sorted":["cs.CL","cs.CV"],"license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","primary_cat":"cs.LG","submitted_at":"2024-10-07T04:00:22Z","title_canon_sha256":"b8d6ebb4d630dd96320c2b223d7089816d271f2c2167363f5568bfeb1b67b43c"},"schema_version":"1.0","source":{"id":"2410.04734","kind":"arxiv","version":2}},"canonical_sha256":"42034fb48a420f25ab8d99ab871965711eee473deba295cced71455a28c84056","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"42034fb48a420f25ab8d99ab871965711eee473deba295cced71455a28c84056","first_computed_at":"2026-07-05T10:19:29.234068Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T10:19:29.234068Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"kh7WA4a1Vl267646aVg5QIryTvvTTSEOG1Xc/HhOxKA4kIpHA2wx395IUYA2bDMFg4MqeLIsfMrieN/ecmUaCw==","signature_status":"signed_v1","signed_at":"2026-07-05T10:19:29.234570Z","signed_message":"canonical_sha256_bytes"},"source_id":"2410.04734","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:e89208672599a0b799c60d8b3895d25dcf4862cba49a59a6bef086b6fe9d8293","sha256:0e605fb15bc23723e14c06b67827091924fb4be8a5c8edfbaa9e008d3da5af68"],"state_sha256":"d69ec3f84e15e35e0f2474920ee53080e023086c035fadaad034cecf1870ad87"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"bTNaEIawtQQOYAeNewoSopqKfLefkoHFT1TCdtR0Co/0pFEiVugdzaboCvj0cKUYc89RfDj5d5WnWW99uu/3DQ==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-04T20:51:16.963636Z","bundle_sha256":"852dcc9978b40ba39dd210b82f142f36bfeb01ffb344666a54b83afda979866a"}}