{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2024:LUPRPSU53RTZLYRLWZI4NMLLT4","short_pith_number":"pith:LUPRPSU5","canonical_record":{"source":{"id":"2411.13697","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2024-11-20T20:28:04Z","cross_cats_sorted":[],"title_canon_sha256":"ee41e900383722b7a2b5530eedd8b9826e5a57a796af774102b7d3f400f30333","abstract_canon_sha256":"73887666de9537bf81f80f79c60dcec60675f5da099e47fe12931154fbf6848a"},"schema_version":"1.0"},"canonical_sha256":"5d1f17ca9ddc6795e22bb651c6b16b9f06507983d6b4deb8976c7790bc737ca0","source":{"kind":"arxiv","id":"2411.13697","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2411.13697","created_at":"2026-07-05T09:38:35Z"},{"alias_kind":"arxiv_version","alias_value":"2411.13697v1","created_at":"2026-07-05T09:38:35Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2411.13697","created_at":"2026-07-05T09:38:35Z"},{"alias_kind":"pith_short_12","alias_value":"LUPRPSU53RTZ","created_at":"2026-07-05T09:38:35Z"},{"alias_kind":"pith_short_16","alias_value":"LUPRPSU53RTZLYRL","created_at":"2026-07-05T09:38:35Z"},{"alias_kind":"pith_short_8","alias_value":"LUPRPSU5","created_at":"2026-07-05T09:38:35Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2024:LUPRPSU53RTZLYRLWZI4NMLLT4","target":"record","payload":{"canonical_record":{"source":{"id":"2411.13697","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2024-11-20T20:28:04Z","cross_cats_sorted":[],"title_canon_sha256":"ee41e900383722b7a2b5530eedd8b9826e5a57a796af774102b7d3f400f30333","abstract_canon_sha256":"73887666de9537bf81f80f79c60dcec60675f5da099e47fe12931154fbf6848a"},"schema_version":"1.0"},"canonical_sha256":"5d1f17ca9ddc6795e22bb651c6b16b9f06507983d6b4deb8976c7790bc737ca0","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:38:35.798848Z","signature_b64":"ad03yTolHH1QvGN159sPDGx6dldVp7IQiI+JF2/AvNtCwZ4BGUbv7w4EA6XKxtMOoLRrRkCAgDZh6lq8f3acDw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"5d1f17ca9ddc6795e22bb651c6b16b9f06507983d6b4deb8976c7790bc737ca0","last_reissued_at":"2026-07-05T09:38:35.798357Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:38:35.798357Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2411.13697","source_version":1,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T09:38:35Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"mTlyfmaDpZQa0qsFqjf6t3592e2JAFKh4naZC3B34KIer1mDsY7gsn4Pss6w2AV4WtvjpLf8iANXrfWJuWVKDw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-12T21:54:30.788234Z"},"content_sha256":"b07ec3d251203de6face438f6ce5246f6810c505fd3495639502129d6c03245c","schema_version":"1.0","event_id":"sha256:b07ec3d251203de6face438f6ce5246f6810c505fd3495639502129d6c03245c"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2024:LUPRPSU53RTZLYRLWZI4NMLLT4","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Decompose and Leverage Preferences from Expert Models for Improving Trustworthiness of MLLMs","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Andreas Vlachos, Michael Schlichtkrull, Rui Cao, Yuming Jiang","submitted_at":"2024-11-20T20:28:04Z","abstract_excerpt":"Multimodal Large Language Models (MLLMs) can enhance trustworthiness by aligning with human preferences. As human preference labeling is laborious, recent works employ evaluation models for assessing MLLMs' responses, using the model-based assessments to automate preference dataset construction. This approach, however, faces challenges with MLLMs' lengthy and compositional responses, which often require diverse reasoning skills that a single evaluation model may not fully possess. Additionally, most existing methods rely on closed-source models as evaluators. To address limitations, we propose"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2411.13697","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/2411.13697/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":"2026-07-05T09:38:35Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"dMPkwwA0P2POSHZpaveHE3YvByI/EmNkyg1dvTNEFfCqsARk/G6D17J3XaIAEgM274/Zn1/z5uzYW/+vpBIfCw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-12T21:54:30.788629Z"},"content_sha256":"3b46b3dee396fc3048ea1eb63e0d0ae80241c0a2ff56fa70ffb5fba7bf1d44d4","schema_version":"1.0","event_id":"sha256:3b46b3dee396fc3048ea1eb63e0d0ae80241c0a2ff56fa70ffb5fba7bf1d44d4"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/LUPRPSU53RTZLYRLWZI4NMLLT4/bundle.json","state_url":"https://pith.science/pith/LUPRPSU53RTZLYRLWZI4NMLLT4/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/LUPRPSU53RTZLYRLWZI4NMLLT4/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":"2026-08-12T21:54:30Z","links":{"resolver":"https://pith.science/pith/LUPRPSU53RTZLYRLWZI4NMLLT4","bundle":"https://pith.science/pith/LUPRPSU53RTZLYRLWZI4NMLLT4/bundle.json","state":"https://pith.science/pith/LUPRPSU53RTZLYRLWZI4NMLLT4/state.json","well_known_bundle":"https://pith.science/.well-known/pith/LUPRPSU53RTZLYRLWZI4NMLLT4/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:LUPRPSU53RTZLYRLWZI4NMLLT4","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":"73887666de9537bf81f80f79c60dcec60675f5da099e47fe12931154fbf6848a","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2024-11-20T20:28:04Z","title_canon_sha256":"ee41e900383722b7a2b5530eedd8b9826e5a57a796af774102b7d3f400f30333"},"schema_version":"1.0","source":{"id":"2411.13697","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2411.13697","created_at":"2026-07-05T09:38:35Z"},{"alias_kind":"arxiv_version","alias_value":"2411.13697v1","created_at":"2026-07-05T09:38:35Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2411.13697","created_at":"2026-07-05T09:38:35Z"},{"alias_kind":"pith_short_12","alias_value":"LUPRPSU53RTZ","created_at":"2026-07-05T09:38:35Z"},{"alias_kind":"pith_short_16","alias_value":"LUPRPSU53RTZLYRL","created_at":"2026-07-05T09:38:35Z"},{"alias_kind":"pith_short_8","alias_value":"LUPRPSU5","created_at":"2026-07-05T09:38:35Z"}],"graph_snapshots":[{"event_id":"sha256:3b46b3dee396fc3048ea1eb63e0d0ae80241c0a2ff56fa70ffb5fba7bf1d44d4","target":"graph","created_at":"2026-07-05T09:38:35Z","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/2411.13697/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Multimodal Large Language Models (MLLMs) can enhance trustworthiness by aligning with human preferences. As human preference labeling is laborious, recent works employ evaluation models for assessing MLLMs' responses, using the model-based assessments to automate preference dataset construction. This approach, however, faces challenges with MLLMs' lengthy and compositional responses, which often require diverse reasoning skills that a single evaluation model may not fully possess. Additionally, most existing methods rely on closed-source models as evaluators. To address limitations, we propose","authors_text":"Andreas Vlachos, Michael Schlichtkrull, Rui Cao, Yuming Jiang","cross_cats":[],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2024-11-20T20:28:04Z","title":"Decompose and Leverage Preferences from Expert Models for Improving Trustworthiness of MLLMs"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2411.13697","kind":"arxiv","version":1},"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:b07ec3d251203de6face438f6ce5246f6810c505fd3495639502129d6c03245c","target":"record","created_at":"2026-07-05T09:38:35Z","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":"73887666de9537bf81f80f79c60dcec60675f5da099e47fe12931154fbf6848a","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2024-11-20T20:28:04Z","title_canon_sha256":"ee41e900383722b7a2b5530eedd8b9826e5a57a796af774102b7d3f400f30333"},"schema_version":"1.0","source":{"id":"2411.13697","kind":"arxiv","version":1}},"canonical_sha256":"5d1f17ca9ddc6795e22bb651c6b16b9f06507983d6b4deb8976c7790bc737ca0","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"5d1f17ca9ddc6795e22bb651c6b16b9f06507983d6b4deb8976c7790bc737ca0","first_computed_at":"2026-07-05T09:38:35.798357Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T09:38:35.798357Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"ad03yTolHH1QvGN159sPDGx6dldVp7IQiI+JF2/AvNtCwZ4BGUbv7w4EA6XKxtMOoLRrRkCAgDZh6lq8f3acDw==","signature_status":"signed_v1","signed_at":"2026-07-05T09:38:35.798848Z","signed_message":"canonical_sha256_bytes"},"source_id":"2411.13697","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:b07ec3d251203de6face438f6ce5246f6810c505fd3495639502129d6c03245c","sha256:3b46b3dee396fc3048ea1eb63e0d0ae80241c0a2ff56fa70ffb5fba7bf1d44d4"],"state_sha256":"8dcf75956a6ef763e7cc516a271fc19b5acf3a2e1674917b82bb893c173f0b15"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"iciAbFyM9eI0Xd8/EkqEyF9D0RiO4T8zba+sn20h5dkT5Rmkb+O55HVe9yva2kBwfMtKJXS2XXg9t/NlUjMlDA==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-12T21:54:30.791192Z","bundle_sha256":"8342f04bb11ee35538d20f061002d5dbd418f4caec0e8687fa5132e225d58423"}}