{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2024:JHTW7AMALJK4JSDR3OR2WJXT3F","short_pith_number":"pith:JHTW7AMA","canonical_record":{"source":{"id":"2402.11411","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2024-02-18T00:56:16Z","cross_cats_sorted":["cs.CL","cs.CV"],"title_canon_sha256":"95495881af9bfc34909c83f89fd32e4e9c83ca55cee19e4cb282e8de90b75049","abstract_canon_sha256":"684c4ab3db146800bb4720a51e3c130124f5c1e1b7bb92be651fce3d29e8a9af"},"schema_version":"1.0"},"canonical_sha256":"49e76f81805a55c4c871dba3ab26f3d953edb4d0e81b5a42139b8ca28b71f008","source":{"kind":"arxiv","id":"2402.11411","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2402.11411","created_at":"2026-05-17T23:38:14Z"},{"alias_kind":"arxiv_version","alias_value":"2402.11411v1","created_at":"2026-05-17T23:38:14Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2402.11411","created_at":"2026-05-17T23:38:14Z"},{"alias_kind":"pith_short_12","alias_value":"JHTW7AMALJK4","created_at":"2026-05-18T12:33:37Z"},{"alias_kind":"pith_short_16","alias_value":"JHTW7AMALJK4JSDR","created_at":"2026-05-18T12:33:37Z"},{"alias_kind":"pith_short_8","alias_value":"JHTW7AMA","created_at":"2026-05-18T12:33:37Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2024:JHTW7AMALJK4JSDR3OR2WJXT3F","target":"record","payload":{"canonical_record":{"source":{"id":"2402.11411","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2024-02-18T00:56:16Z","cross_cats_sorted":["cs.CL","cs.CV"],"title_canon_sha256":"95495881af9bfc34909c83f89fd32e4e9c83ca55cee19e4cb282e8de90b75049","abstract_canon_sha256":"684c4ab3db146800bb4720a51e3c130124f5c1e1b7bb92be651fce3d29e8a9af"},"schema_version":"1.0"},"canonical_sha256":"49e76f81805a55c4c871dba3ab26f3d953edb4d0e81b5a42139b8ca28b71f008","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-17T23:38:14.301978Z","signature_b64":"lJwLq7cAoEFNyhF3BP/8U+zA4IL/j5dUzks3bapVbt7RM+FZvbpq+GFeWV09NgW7E8z3wkdgTqjmamnFckLOBg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"49e76f81805a55c4c871dba3ab26f3d953edb4d0e81b5a42139b8ca28b71f008","last_reissued_at":"2026-05-17T23:38:14.301452Z","signature_status":"signed_v1","first_computed_at":"2026-05-17T23:38:14.301452Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2402.11411","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":"2026-05-17T23:38:14Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"JeirgnA+euD9r33YB3sjotGQA0pKI85HgcMSN8rOOSXzIz9dhC8BPAJmovjPmCmZJLuiR6yelLULJ96JiVBXCA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-23T04:54:47.911399Z"},"content_sha256":"79562c7d890010a97181de90fef81d2844a991bf506a3c4899b31747444fa4b6","schema_version":"1.0","event_id":"sha256:79562c7d890010a97181de90fef81d2844a991bf506a3c4899b31747444fa4b6"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2024:JHTW7AMALJK4JSDR3OR2WJXT3F","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Aligning Modalities in Vision Large Language Models via Preference Fine-tuning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"Preference fine-tuning on automatically generated hallucinated responses aligns vision and language modalities in large models while cutting hallucinations.","cross_cats":["cs.CL","cs.CV"],"primary_cat":"cs.LG","authors_text":"Chelsea Finn, Chenhang Cui, Huaxiu Yao, Rafael Rafailov, Yiyang Zhou","submitted_at":"2024-02-18T00:56:16Z","abstract_excerpt":"Instruction-following Vision Large Language Models (VLLMs) have achieved significant progress recently on a variety of tasks. These approaches merge strong pre-trained vision models and large language models (LLMs). Since these components are trained separately, the learned representations need to be aligned with joint training on additional image-language pairs. This procedure is not perfect and can cause the model to hallucinate - provide answers that do not accurately reflect the image, even when the core LLM is highly factual and the vision backbone has sufficiently complete representation"},"claims":{"count":4,"items":[{"kind":"strongest_claim","text":"In experiments across broad benchmarks, we show that we can not only reduce hallucinations, but improve model performance across standard benchmarks, outperforming prior approaches.","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","text":"The two-stage automated generation of dispreferred responses (GPT-4V hallucination injection and image distortion) produces high-quality preference pairs that accurately reflect and correct the model's hallucination behavior when used in DPO.","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","text":"POVID generates AI-created preference data to fine-tune vision-language models with DPO, reducing hallucinations and improving benchmark scores.","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"},{"kind":"headline","text":"Preference fine-tuning on automatically generated hallucinated responses aligns vision and language modalities in large models while cutting hallucinations.","source":"verdict.pith_extraction.headline","status":"machine_extracted","claim_id":"C4","attestation":"unclaimed"}],"snapshot_sha256":"fc41d789a43373108e2b1e47bfe0168f29991977cfe7ec321abbbf203a40db8a"},"source":{"id":"2402.11411","kind":"arxiv","version":1},"verdict":{"id":"c94e6e5a-3650-40e1-8bb0-efed02110008","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-17T10:54:31.659670Z","strongest_claim":"In experiments across broad benchmarks, we show that we can not only reduce hallucinations, but improve model performance across standard benchmarks, outperforming prior approaches.","one_line_summary":"POVID generates AI-created preference data to fine-tune vision-language models with DPO, reducing hallucinations and improving benchmark scores.","pipeline_version":"pith-pipeline@v0.9.0","weakest_assumption":"The two-stage automated generation of dispreferred responses (GPT-4V hallucination injection and image distortion) produces high-quality preference pairs that accurately reflect and correct the model's hallucination behavior when used in DPO.","pith_extraction_headline":"Preference fine-tuning on automatically generated hallucinated responses aligns vision and language modalities in large models while cutting hallucinations."},"references":{"count":155,"sample":[{"doi":"","year":2023,"title":"InstructBLIP: Towards General-purpose Vision-Language Models with Instruction Tuning , author=. 2023 , eprint=","work_id":"2a233147-17ae-4a84-a51a-aeaa6303e8e1","ref_index":18,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2000,"title":"Langley , title =","work_id":"6cd283dc-0548-45e9-af07-6bc1005593ad","ref_index":20,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":1980,"title":"T. M. Mitchell. The Need for Biases in Learning Generalizations. 1980","work_id":"6b09bca6-ef5d-4a83-8c8e-219f23cbd761","ref_index":21,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":null,"title":"M. J. Kearns , title =","work_id":"8efd8073-6f5d-45c5-94d5-62d366b52518","ref_index":22,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":1983,"title":"Machine Learning: An Artificial Intelligence Approach, Vol. I. 1983","work_id":"51835800-f16e-4534-8339-d3ea09147556","ref_index":23,"cited_arxiv_id":"","is_internal_anchor":false}],"resolved_work":155,"snapshot_sha256":"5b1f65218dc218aa0a2bbc2f22c81feb7663e2379061bb550038898361d7e30c","internal_anchors":22},"formal_canon":{"evidence_count":2,"snapshot_sha256":"41ded6d331100e664479d2b235b3d8c2d8b5b1aa5b27fbfa16f9871c6754a422"},"author_claims":{"count":0,"strong_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"builder_version":"pith-number-builder-2026-05-17-v1"},"verdict_id":"c94e6e5a-3650-40e1-8bb0-efed02110008"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-05-17T23:38:14Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"VeA8JiRo41RHNNv7ioyDHKU68ghpv6LKM4J5yrGN4IAl/ijsW6LR3k9WMCMy8Jukn+B5HrSKsltw6LCdWti9BA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-23T04:54:47.912423Z"},"content_sha256":"1a673c5570d9ebca5987e1f322956d5406adad6bcfd6da3982f238801dab1975","schema_version":"1.0","event_id":"sha256:1a673c5570d9ebca5987e1f322956d5406adad6bcfd6da3982f238801dab1975"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/JHTW7AMALJK4JSDR3OR2WJXT3F/bundle.json","state_url":"https://pith.science/pith/JHTW7AMALJK4JSDR3OR2WJXT3F/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/JHTW7AMALJK4JSDR3OR2WJXT3F/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":"2026-05-23T04:54:47Z","links":{"resolver":"https://pith.science/pith/JHTW7AMALJK4JSDR3OR2WJXT3F","bundle":"https://pith.science/pith/JHTW7AMALJK4JSDR3OR2WJXT3F/bundle.json","state":"https://pith.science/pith/JHTW7AMALJK4JSDR3OR2WJXT3F/state.json","well_known_bundle":"https://pith.science/.well-known/pith/JHTW7AMALJK4JSDR3OR2WJXT3F/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:JHTW7AMALJK4JSDR3OR2WJXT3F","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":"684c4ab3db146800bb4720a51e3c130124f5c1e1b7bb92be651fce3d29e8a9af","cross_cats_sorted":["cs.CL","cs.CV"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2024-02-18T00:56:16Z","title_canon_sha256":"95495881af9bfc34909c83f89fd32e4e9c83ca55cee19e4cb282e8de90b75049"},"schema_version":"1.0","source":{"id":"2402.11411","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2402.11411","created_at":"2026-05-17T23:38:14Z"},{"alias_kind":"arxiv_version","alias_value":"2402.11411v1","created_at":"2026-05-17T23:38:14Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2402.11411","created_at":"2026-05-17T23:38:14Z"},{"alias_kind":"pith_short_12","alias_value":"JHTW7AMALJK4","created_at":"2026-05-18T12:33:37Z"},{"alias_kind":"pith_short_16","alias_value":"JHTW7AMALJK4JSDR","created_at":"2026-05-18T12:33:37Z"},{"alias_kind":"pith_short_8","alias_value":"JHTW7AMA","created_at":"2026-05-18T12:33:37Z"}],"graph_snapshots":[{"event_id":"sha256:1a673c5570d9ebca5987e1f322956d5406adad6bcfd6da3982f238801dab1975","target":"graph","created_at":"2026-05-17T23:38:14Z","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":4,"items":[{"attestation":"unclaimed","claim_id":"C1","kind":"strongest_claim","source":"verdict.strongest_claim","status":"machine_extracted","text":"In experiments across broad benchmarks, we show that we can not only reduce hallucinations, but improve model performance across standard benchmarks, outperforming prior approaches."},{"attestation":"unclaimed","claim_id":"C2","kind":"weakest_assumption","source":"verdict.weakest_assumption","status":"machine_extracted","text":"The two-stage automated generation of dispreferred responses (GPT-4V hallucination injection and image distortion) produces high-quality preference pairs that accurately reflect and correct the model's hallucination behavior when used in DPO."},{"attestation":"unclaimed","claim_id":"C3","kind":"one_line_summary","source":"verdict.one_line_summary","status":"machine_extracted","text":"POVID generates AI-created preference data to fine-tune vision-language models with DPO, reducing hallucinations and improving benchmark scores."},{"attestation":"unclaimed","claim_id":"C4","kind":"headline","source":"verdict.pith_extraction.headline","status":"machine_extracted","text":"Preference fine-tuning on automatically generated hallucinated responses aligns vision and language modalities in large models while cutting hallucinations."}],"snapshot_sha256":"fc41d789a43373108e2b1e47bfe0168f29991977cfe7ec321abbbf203a40db8a"},"formal_canon":{"evidence_count":2,"snapshot_sha256":"41ded6d331100e664479d2b235b3d8c2d8b5b1aa5b27fbfa16f9871c6754a422"},"paper":{"abstract_excerpt":"Instruction-following Vision Large Language Models (VLLMs) have achieved significant progress recently on a variety of tasks. These approaches merge strong pre-trained vision models and large language models (LLMs). Since these components are trained separately, the learned representations need to be aligned with joint training on additional image-language pairs. This procedure is not perfect and can cause the model to hallucinate - provide answers that do not accurately reflect the image, even when the core LLM is highly factual and the vision backbone has sufficiently complete representation","authors_text":"Chelsea Finn, Chenhang Cui, Huaxiu Yao, Rafael Rafailov, Yiyang Zhou","cross_cats":["cs.CL","cs.CV"],"headline":"Preference fine-tuning on automatically generated hallucinated responses aligns vision and language modalities in large models while cutting hallucinations.","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2024-02-18T00:56:16Z","title":"Aligning Modalities in Vision Large Language Models via Preference Fine-tuning"},"references":{"count":155,"internal_anchors":22,"resolved_work":155,"sample":[{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":18,"title":"InstructBLIP: Towards General-purpose Vision-Language Models with Instruction Tuning , author=. 2023 , eprint=","work_id":"2a233147-17ae-4a84-a51a-aeaa6303e8e1","year":2023},{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":20,"title":"Langley , title =","work_id":"6cd283dc-0548-45e9-af07-6bc1005593ad","year":2000},{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":21,"title":"T. M. Mitchell. The Need for Biases in Learning Generalizations. 1980","work_id":"6b09bca6-ef5d-4a83-8c8e-219f23cbd761","year":1980},{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":22,"title":"M. J. Kearns , title =","work_id":"8efd8073-6f5d-45c5-94d5-62d366b52518","year":null},{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":23,"title":"Machine Learning: An Artificial Intelligence Approach, Vol. I. 1983","work_id":"51835800-f16e-4534-8339-d3ea09147556","year":1983}],"snapshot_sha256":"5b1f65218dc218aa0a2bbc2f22c81feb7663e2379061bb550038898361d7e30c"},"source":{"id":"2402.11411","kind":"arxiv","version":1},"verdict":{"created_at":"2026-05-17T10:54:31.659670Z","id":"c94e6e5a-3650-40e1-8bb0-efed02110008","model_set":{"reader":"grok-4.3"},"one_line_summary":"POVID generates AI-created preference data to fine-tune vision-language models with DPO, reducing hallucinations and improving benchmark scores.","pipeline_version":"pith-pipeline@v0.9.0","pith_extraction_headline":"Preference fine-tuning on automatically generated hallucinated responses aligns vision and language modalities in large models while cutting hallucinations.","strongest_claim":"In experiments across broad benchmarks, we show that we can not only reduce hallucinations, but improve model performance across standard benchmarks, outperforming prior approaches.","weakest_assumption":"The two-stage automated generation of dispreferred responses (GPT-4V hallucination injection and image distortion) produces high-quality preference pairs that accurately reflect and correct the model's hallucination behavior when used in DPO."}},"verdict_id":"c94e6e5a-3650-40e1-8bb0-efed02110008"}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:79562c7d890010a97181de90fef81d2844a991bf506a3c4899b31747444fa4b6","target":"record","created_at":"2026-05-17T23:38:14Z","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":"684c4ab3db146800bb4720a51e3c130124f5c1e1b7bb92be651fce3d29e8a9af","cross_cats_sorted":["cs.CL","cs.CV"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2024-02-18T00:56:16Z","title_canon_sha256":"95495881af9bfc34909c83f89fd32e4e9c83ca55cee19e4cb282e8de90b75049"},"schema_version":"1.0","source":{"id":"2402.11411","kind":"arxiv","version":1}},"canonical_sha256":"49e76f81805a55c4c871dba3ab26f3d953edb4d0e81b5a42139b8ca28b71f008","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"49e76f81805a55c4c871dba3ab26f3d953edb4d0e81b5a42139b8ca28b71f008","first_computed_at":"2026-05-17T23:38:14.301452Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-17T23:38:14.301452Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"lJwLq7cAoEFNyhF3BP/8U+zA4IL/j5dUzks3bapVbt7RM+FZvbpq+GFeWV09NgW7E8z3wkdgTqjmamnFckLOBg==","signature_status":"signed_v1","signed_at":"2026-05-17T23:38:14.301978Z","signed_message":"canonical_sha256_bytes"},"source_id":"2402.11411","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:79562c7d890010a97181de90fef81d2844a991bf506a3c4899b31747444fa4b6","sha256:1a673c5570d9ebca5987e1f322956d5406adad6bcfd6da3982f238801dab1975"],"state_sha256":"79b83da5bd732fba682f39e089fba10f314acee6ccce11d5c7cf9b0224575015"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"WPVjQDLWTOJo7CRLisFLjEquopv23RTLo99jBi/zFNBt7dMs1NRKa+EhxIDo/TmprLiVDAT+l9oJEzVl0zjPAA==","signed_message":"bundle_sha256_bytes","signed_at":"2026-05-23T04:54:47.916236Z","bundle_sha256":"9318ef86e09b67e45bc8361b2e0cfaf29adb59e7b90b0659ee16e8b0509cceb4"}}