{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:C4CRBQ34B7PI7KTYNJINGO6LSU","short_pith_number":"pith:C4CRBQ34","schema_version":"1.0","canonical_sha256":"170510c37c0fde8faa786a50d33bcb9526c1299b373585b7fb040c7b02cf4af1","source":{"kind":"arxiv","id":"2403.00425","version":2},"attestation_state":"computed","paper":{"title":"HALC: Object Hallucination Reduction via Adaptive Focal-Contrast Decoding","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.LG"],"primary_cat":"cs.CV","authors_text":"Bo Li, Hongyin Luo, Huaxiu Yao, Jiawei Zhou, Zhaorun Chen, Zhuokai Zhao","submitted_at":"2024-03-01T10:21:52Z","abstract_excerpt":"While large vision-language models (LVLMs) have demonstrated impressive capabilities in interpreting multi-modal contexts, they invariably suffer from object hallucinations (OH). We introduce HALC, a novel decoding algorithm designed to mitigate OH in LVLMs. HALC leverages distinct fine-grained optimal visual information in vision-language tasks and operates on both local and global contexts simultaneously. Specifically, HALC integrates a robust auto-focal grounding mechanism (locally) to correct hallucinated tokens on the fly, and a specialized beam search algorithm (globally) to significantl"},"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":"2403.00425","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2024-03-01T10:21:52Z","cross_cats_sorted":["cs.AI","cs.LG"],"title_canon_sha256":"9e8d0551a3d22a416254e286fdea4df2d9ccfd9bcd75492811f0bd8dd4c46306","abstract_canon_sha256":"1ee577e9f670692b09b70cfd0af49efc1f58c96462238536cff79fb2ed0b2699"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T08:29:23.485054Z","signature_b64":"0EEg6c3FjS5SVBtcKqarPMxw7o5359S6Qj7Nswyf/RiaDbaKuk3B81CTM2dP64OPXh0mJFuDheNZkguTsRDhBw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"170510c37c0fde8faa786a50d33bcb9526c1299b373585b7fb040c7b02cf4af1","last_reissued_at":"2026-07-05T08:29:23.484579Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T08:29:23.484579Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"HALC: Object Hallucination Reduction via Adaptive Focal-Contrast Decoding","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.LG"],"primary_cat":"cs.CV","authors_text":"Bo Li, Hongyin Luo, Huaxiu Yao, Jiawei Zhou, Zhaorun Chen, Zhuokai Zhao","submitted_at":"2024-03-01T10:21:52Z","abstract_excerpt":"While large vision-language models (LVLMs) have demonstrated impressive capabilities in interpreting multi-modal contexts, they invariably suffer from object hallucinations (OH). We introduce HALC, a novel decoding algorithm designed to mitigate OH in LVLMs. HALC leverages distinct fine-grained optimal visual information in vision-language tasks and operates on both local and global contexts simultaneously. Specifically, HALC integrates a robust auto-focal grounding mechanism (locally) to correct hallucinated tokens on the fly, and a specialized beam search algorithm (globally) to significantl"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2403.00425","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/2403.00425/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":"2403.00425","created_at":"2026-07-05T08:29:23.484642+00:00"},{"alias_kind":"arxiv_version","alias_value":"2403.00425v2","created_at":"2026-07-05T08:29:23.484642+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2403.00425","created_at":"2026-07-05T08:29:23.484642+00:00"},{"alias_kind":"pith_short_12","alias_value":"C4CRBQ34B7PI","created_at":"2026-07-05T08:29:23.484642+00:00"},{"alias_kind":"pith_short_16","alias_value":"C4CRBQ34B7PI7KTY","created_at":"2026-07-05T08:29:23.484642+00:00"},{"alias_kind":"pith_short_8","alias_value":"C4CRBQ34","created_at":"2026-07-05T08:29:23.484642+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":12,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.27596","citing_title":"Dismantling Pathological Shortcuts: A Causal Framework for Faithful LVLM Decoding","ref_index":6,"is_internal_anchor":false},{"citing_arxiv_id":"2605.24602","citing_title":"Correcting Visual Blur Induced by Attention Distraction to Reduce Hallucinations: Algorithm and Theory","ref_index":5,"is_internal_anchor":false},{"citing_arxiv_id":"2605.24957","citing_title":"Mitigating Object Hallucinations in Vision-Language Models through Region-Aware Attention Recalibration","ref_index":8,"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":4,"is_internal_anchor":false},{"citing_arxiv_id":"2606.09859","citing_title":"Mitigating Manifold Departure: Uncertainty-Aware Subspace Rectification for Trustworthy MLLM Decoding","ref_index":24,"is_internal_anchor":false},{"citing_arxiv_id":"2507.12455","citing_title":"Mitigating Object Hallucinations via Sentence-Level Early Intervention","ref_index":6,"is_internal_anchor":false},{"citing_arxiv_id":"2505.13255","citing_title":"Policy Contrastive Decoding for Robotic Foundation Models","ref_index":6,"is_internal_anchor":false},{"citing_arxiv_id":"2605.10676","citing_title":"Not Blind but Silenced: Rebalancing Vision and Language via Adversarial Counter-Commonsense Equilibrium","ref_index":35,"is_internal_anchor":false},{"citing_arxiv_id":"2605.09443","citing_title":"Through the Lens of Character: Resolving Modality-Role Interference in Multimodal Role-Playing Agent","ref_index":37,"is_internal_anchor":false},{"citing_arxiv_id":"2404.18930","citing_title":"Hallucination of Multimodal Large Language Models: A Survey","ref_index":31,"is_internal_anchor":false},{"citing_arxiv_id":"2604.21027","citing_title":"HypEHR: Hyperbolic Modeling of Electronic Health Records for Efficient Question Answering","ref_index":266,"is_internal_anchor":false},{"citing_arxiv_id":"2605.04641","citing_title":"CAST: Mitigating Object Hallucination in Large Vision-Language Models via Caption-Guided Visual Attention Steering","ref_index":46,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/C4CRBQ34B7PI7KTYNJINGO6LSU","json":"https://pith.science/pith/C4CRBQ34B7PI7KTYNJINGO6LSU.json","graph_json":"https://pith.science/api/pith-number/C4CRBQ34B7PI7KTYNJINGO6LSU/graph.json","events_json":"https://pith.science/api/pith-number/C4CRBQ34B7PI7KTYNJINGO6LSU/events.json","paper":"https://pith.science/paper/C4CRBQ34"},"agent_actions":{"view_html":"https://pith.science/pith/C4CRBQ34B7PI7KTYNJINGO6LSU","download_json":"https://pith.science/pith/C4CRBQ34B7PI7KTYNJINGO6LSU.json","view_paper":"https://pith.science/paper/C4CRBQ34","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2403.00425&json=true","fetch_graph":"https://pith.science/api/pith-number/C4CRBQ34B7PI7KTYNJINGO6LSU/graph.json","fetch_events":"https://pith.science/api/pith-number/C4CRBQ34B7PI7KTYNJINGO6LSU/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/C4CRBQ34B7PI7KTYNJINGO6LSU/action/timestamp_anchor","attest_storage":"https://pith.science/pith/C4CRBQ34B7PI7KTYNJINGO6LSU/action/storage_attestation","attest_author":"https://pith.science/pith/C4CRBQ34B7PI7KTYNJINGO6LSU/action/author_attestation","sign_citation":"https://pith.science/pith/C4CRBQ34B7PI7KTYNJINGO6LSU/action/citation_signature","submit_replication":"https://pith.science/pith/C4CRBQ34B7PI7KTYNJINGO6LSU/action/replication_record"}},"created_at":"2026-07-05T08:29:23.484642+00:00","updated_at":"2026-07-05T08:29:23.484642+00:00"}