{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:EPS3EX6MXKZHMLHB4DC3MODSBV","short_pith_number":"pith:EPS3EX6M","schema_version":"1.0","canonical_sha256":"23e5b25fccbab2762ce1e0c5b638720d69cfd2e5b5972fe4d977f62fb4235eaa","source":{"kind":"arxiv","id":"2408.05337","version":1},"attestation_state":"computed","paper":{"title":"VACoDe: Visual Augmented Contrastive Decoding","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CV","authors_text":"Boryeong Cho, Sangmin Bae, Se-Young Yun, Sihyeon Kim, Sumyeong Ahn","submitted_at":"2024-07-26T15:49:31Z","abstract_excerpt":"Despite the astonishing performance of recent Large Vision-Language Models (LVLMs), these models often generate inaccurate responses. To address this issue, previous studies have focused on mitigating hallucinations by employing contrastive decoding (CD) with augmented images, which amplifies the contrast with the original image. However, these methods have limitations, including reliance on a single augmentation, which is restrictive for certain tasks, as well as the high cost of using external knowledge. In this study, we address these limitations by exploring how to utilize multiple image a"},"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":"2408.05337","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2024-07-26T15:49:31Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"f750251b4f2858d7a3709e9366f6bda78b51b2aae6e77bfea5bb3bc57d72b78d","abstract_canon_sha256":"a2a930282647919232cd0eb15f1d71821807910624b664b6887ed4fd54c32eaa"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T08:54:18.805612Z","signature_b64":"iXS97dtcAOCWi/VVcSFksjxaW64EiD7h4dUJYbX0lbUY6HGcEkVRGkyQSi28/oJt0ee/4CJHFOSH4Pwx2jlsDA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"23e5b25fccbab2762ce1e0c5b638720d69cfd2e5b5972fe4d977f62fb4235eaa","last_reissued_at":"2026-07-05T08:54:18.805121Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T08:54:18.805121Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"VACoDe: Visual Augmented Contrastive Decoding","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CV","authors_text":"Boryeong Cho, Sangmin Bae, Se-Young Yun, Sihyeon Kim, Sumyeong Ahn","submitted_at":"2024-07-26T15:49:31Z","abstract_excerpt":"Despite the astonishing performance of recent Large Vision-Language Models (LVLMs), these models often generate inaccurate responses. To address this issue, previous studies have focused on mitigating hallucinations by employing contrastive decoding (CD) with augmented images, which amplifies the contrast with the original image. However, these methods have limitations, including reliance on a single augmentation, which is restrictive for certain tasks, as well as the high cost of using external knowledge. In this study, we address these limitations by exploring how to utilize multiple image a"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2408.05337","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/2408.05337/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":"2408.05337","created_at":"2026-07-05T08:54:18.805182+00:00"},{"alias_kind":"arxiv_version","alias_value":"2408.05337v1","created_at":"2026-07-05T08:54:18.805182+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2408.05337","created_at":"2026-07-05T08:54:18.805182+00:00"},{"alias_kind":"pith_short_12","alias_value":"EPS3EX6MXKZH","created_at":"2026-07-05T08:54:18.805182+00:00"},{"alias_kind":"pith_short_16","alias_value":"EPS3EX6MXKZHMLHB","created_at":"2026-07-05T08:54:18.805182+00:00"},{"alias_kind":"pith_short_8","alias_value":"EPS3EX6M","created_at":"2026-07-05T08:54:18.805182+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":4,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2607.00247","citing_title":"Adaptive Perturbation Selection for Contrastive Audio Decoding","ref_index":21,"is_internal_anchor":false},{"citing_arxiv_id":"2605.31429","citing_title":"YARD: Y-Architecture Register Decoding for Efficient Hallucination Mitigation in Large Vision-Language Models","ref_index":20,"is_internal_anchor":false},{"citing_arxiv_id":"2605.23344","citing_title":"CHASD: Language Increment-Calibrated Contrastive Decoding against Hallucination in LVLMs","ref_index":17,"is_internal_anchor":false},{"citing_arxiv_id":"2404.18930","citing_title":"Hallucination of Multimodal Large Language Models: A Survey","ref_index":90,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/EPS3EX6MXKZHMLHB4DC3MODSBV","json":"https://pith.science/pith/EPS3EX6MXKZHMLHB4DC3MODSBV.json","graph_json":"https://pith.science/api/pith-number/EPS3EX6MXKZHMLHB4DC3MODSBV/graph.json","events_json":"https://pith.science/api/pith-number/EPS3EX6MXKZHMLHB4DC3MODSBV/events.json","paper":"https://pith.science/paper/EPS3EX6M"},"agent_actions":{"view_html":"https://pith.science/pith/EPS3EX6MXKZHMLHB4DC3MODSBV","download_json":"https://pith.science/pith/EPS3EX6MXKZHMLHB4DC3MODSBV.json","view_paper":"https://pith.science/paper/EPS3EX6M","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2408.05337&json=true","fetch_graph":"https://pith.science/api/pith-number/EPS3EX6MXKZHMLHB4DC3MODSBV/graph.json","fetch_events":"https://pith.science/api/pith-number/EPS3EX6MXKZHMLHB4DC3MODSBV/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/EPS3EX6MXKZHMLHB4DC3MODSBV/action/timestamp_anchor","attest_storage":"https://pith.science/pith/EPS3EX6MXKZHMLHB4DC3MODSBV/action/storage_attestation","attest_author":"https://pith.science/pith/EPS3EX6MXKZHMLHB4DC3MODSBV/action/author_attestation","sign_citation":"https://pith.science/pith/EPS3EX6MXKZHMLHB4DC3MODSBV/action/citation_signature","submit_replication":"https://pith.science/pith/EPS3EX6MXKZHMLHB4DC3MODSBV/action/replication_record"}},"created_at":"2026-07-05T08:54:18.805182+00:00","updated_at":"2026-07-05T08:54:18.805182+00:00"}