{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:O3KYTM2PE32ROEQUSKOZ4CLMTH","short_pith_number":"pith:O3KYTM2P","schema_version":"1.0","canonical_sha256":"76d589b34f26f5171214929d9e096c99d8b9715ab69ec318c7ae3ce35aa13305","source":{"kind":"arxiv","id":"2502.06130","version":2},"attestation_state":"computed","paper":{"title":"Self-Correcting Decoding with Generative Feedback for Mitigating Hallucinations in Large Vision-Language Models","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CL"],"primary_cat":"cs.CV","authors_text":"Ce Zhang, Deva Ramanan, Katia Sycara, Louis-Philippe Morency, Martin Q. Ma, Russ Salakhutdinov, Simon Stepputtis, Yaqi Xie, Zhehan Kan, Zifu Wan","submitted_at":"2025-02-10T03:43:55Z","abstract_excerpt":"While recent Large Vision-Language Models (LVLMs) have shown remarkable performance in multi-modal tasks, they are prone to generating hallucinatory text responses that do not align with the given visual input, which restricts their practical applicability in real-world scenarios. In this work, inspired by the observation that the text-to-image generation process is the inverse of image-conditioned response generation in LVLMs, we explore the potential of leveraging text-to-image generative models to assist in mitigating hallucinations in LVLMs. We discover that generative models can offer val"},"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":"2502.06130","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2025-02-10T03:43:55Z","cross_cats_sorted":["cs.CL"],"title_canon_sha256":"f4822e30c4059ebcc11532992bef15b61173200973f7097be8d855c028f37061","abstract_canon_sha256":"6881a414d937c5749118dfe788d79c928c89c4fcd2bfc3bf8fe24eb5e6a0aac1"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T12:07:58.353324Z","signature_b64":"RVRpemKnkvmFXjrNLjek3YAvwnfs9QNi5XlLypIFXX+3bUGYEPaKadlU8tGKkjTnocyGgcDwHXHbXjwFAEaLAw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"76d589b34f26f5171214929d9e096c99d8b9715ab69ec318c7ae3ce35aa13305","last_reissued_at":"2026-07-05T12:07:58.352749Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T12:07:58.352749Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Self-Correcting Decoding with Generative Feedback for Mitigating Hallucinations in Large Vision-Language Models","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CL"],"primary_cat":"cs.CV","authors_text":"Ce Zhang, Deva Ramanan, Katia Sycara, Louis-Philippe Morency, Martin Q. Ma, Russ Salakhutdinov, Simon Stepputtis, Yaqi Xie, Zhehan Kan, Zifu Wan","submitted_at":"2025-02-10T03:43:55Z","abstract_excerpt":"While recent Large Vision-Language Models (LVLMs) have shown remarkable performance in multi-modal tasks, they are prone to generating hallucinatory text responses that do not align with the given visual input, which restricts their practical applicability in real-world scenarios. In this work, inspired by the observation that the text-to-image generation process is the inverse of image-conditioned response generation in LVLMs, we explore the potential of leveraging text-to-image generative models to assist in mitigating hallucinations in LVLMs. We discover that generative models can offer val"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2502.06130","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/2502.06130/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":"2502.06130","created_at":"2026-07-05T12:07:58.352813+00:00"},{"alias_kind":"arxiv_version","alias_value":"2502.06130v2","created_at":"2026-07-05T12:07:58.352813+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2502.06130","created_at":"2026-07-05T12:07:58.352813+00:00"},{"alias_kind":"pith_short_12","alias_value":"O3KYTM2PE32R","created_at":"2026-07-05T12:07:58.352813+00:00"},{"alias_kind":"pith_short_16","alias_value":"O3KYTM2PE32ROEQU","created_at":"2026-07-05T12:07:58.352813+00:00"},{"alias_kind":"pith_short_8","alias_value":"O3KYTM2P","created_at":"2026-07-05T12:07:58.352813+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":7,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.27596","citing_title":"Dismantling Pathological Shortcuts: A Causal Framework for Faithful LVLM Decoding","ref_index":37,"is_internal_anchor":false},{"citing_arxiv_id":"2606.31054","citing_title":"ADAPT: Attention Dynamics Alignment with Preference Tuning for Faithful MLLMs","ref_index":40,"is_internal_anchor":false},{"citing_arxiv_id":"2605.04874","citing_title":"Uncertainty-Aware Exploratory Direct Preference Optimization for Multimodal Large Language Models","ref_index":42,"is_internal_anchor":false},{"citing_arxiv_id":"2404.18930","citing_title":"Hallucination of Multimodal Large Language Models: A Survey","ref_index":206,"is_internal_anchor":false},{"citing_arxiv_id":"2604.12424","citing_title":"Decoding by Perturbation: Mitigating MLLM Hallucinations via Dynamic Textual Perturbation","ref_index":55,"is_internal_anchor":false},{"citing_arxiv_id":"2604.20366","citing_title":"Mitigating Hallucinations in Large Vision-Language Models without Performance Degradation","ref_index":161,"is_internal_anchor":false},{"citing_arxiv_id":"2604.21027","citing_title":"HypEHR: Hyperbolic Modeling of Electronic Health Records for Efficient Question Answering","ref_index":282,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/O3KYTM2PE32ROEQUSKOZ4CLMTH","json":"https://pith.science/pith/O3KYTM2PE32ROEQUSKOZ4CLMTH.json","graph_json":"https://pith.science/api/pith-number/O3KYTM2PE32ROEQUSKOZ4CLMTH/graph.json","events_json":"https://pith.science/api/pith-number/O3KYTM2PE32ROEQUSKOZ4CLMTH/events.json","paper":"https://pith.science/paper/O3KYTM2P"},"agent_actions":{"view_html":"https://pith.science/pith/O3KYTM2PE32ROEQUSKOZ4CLMTH","download_json":"https://pith.science/pith/O3KYTM2PE32ROEQUSKOZ4CLMTH.json","view_paper":"https://pith.science/paper/O3KYTM2P","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2502.06130&json=true","fetch_graph":"https://pith.science/api/pith-number/O3KYTM2PE32ROEQUSKOZ4CLMTH/graph.json","fetch_events":"https://pith.science/api/pith-number/O3KYTM2PE32ROEQUSKOZ4CLMTH/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/O3KYTM2PE32ROEQUSKOZ4CLMTH/action/timestamp_anchor","attest_storage":"https://pith.science/pith/O3KYTM2PE32ROEQUSKOZ4CLMTH/action/storage_attestation","attest_author":"https://pith.science/pith/O3KYTM2PE32ROEQUSKOZ4CLMTH/action/author_attestation","sign_citation":"https://pith.science/pith/O3KYTM2PE32ROEQUSKOZ4CLMTH/action/citation_signature","submit_replication":"https://pith.science/pith/O3KYTM2PE32ROEQUSKOZ4CLMTH/action/replication_record"}},"created_at":"2026-07-05T12:07:58.352813+00:00","updated_at":"2026-07-05T12:07:58.352813+00:00"}