{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:RPZKZWRIX4KJ4EVAYHN4WW5RHO","short_pith_number":"pith:RPZKZWRI","schema_version":"1.0","canonical_sha256":"8bf2acda28bf149e12a0c1dbcb5bb13ba74c19b60e02835b2af08c3cd06502ab","source":{"kind":"arxiv","id":"2605.21300","version":1},"attestation_state":"computed","paper":{"title":"Reducing Object Hallucination in LVLMs via Emphasizing Image-negative Tokens","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Deepu Rajan, Meng Shen, Minghao Wu","submitted_at":"2026-05-20T15:29:20Z","abstract_excerpt":"Object hallucination is a significant challenge that hinders the application of large vision-language models (LVLMs) in practice. We hypothesize that one possible origin of hallucination is the model's tendency to prioritize text generation over meaningful interaction with images. To explore this, we examine the generation process and categorize text tokens into three groups: image-positive, invariant, and negative, based on their visual dependence on input image tokens. Our analysis reveals that most generated tokens are minimally influenced by the image information. This suggests that during"},"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":"2605.21300","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2026-05-20T15:29:20Z","cross_cats_sorted":[],"title_canon_sha256":"f8ad704949eed5fee968cd6ea40f1314cd33fda93c06c49471e40d6e85ee2ea9","abstract_canon_sha256":"a4ea61e9be1cdebdcb01f4b98276d49344deddabaeda6d119636530037bf3dfa"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-21T02:05:27.955597Z","signature_b64":"9QS2NpiZ+utuKRT22ytr5v3AC8Uhdjg7dSV3rzRYMphtPr18pHEq9CT453OEn1G8GmsZMB8Bb1CdWiewUC9BCQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"8bf2acda28bf149e12a0c1dbcb5bb13ba74c19b60e02835b2af08c3cd06502ab","last_reissued_at":"2026-05-21T02:05:27.954844Z","signature_status":"signed_v1","first_computed_at":"2026-05-21T02:05:27.954844Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Reducing Object Hallucination in LVLMs via Emphasizing Image-negative Tokens","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Deepu Rajan, Meng Shen, Minghao Wu","submitted_at":"2026-05-20T15:29:20Z","abstract_excerpt":"Object hallucination is a significant challenge that hinders the application of large vision-language models (LVLMs) in practice. We hypothesize that one possible origin of hallucination is the model's tendency to prioritize text generation over meaningful interaction with images. To explore this, we examine the generation process and categorize text tokens into three groups: image-positive, invariant, and negative, based on their visual dependence on input image tokens. Our analysis reveals that most generated tokens are minimally influenced by the image information. This suggests that during"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2605.21300","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/2605.21300/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":"2605.21300","created_at":"2026-05-21T02:05:27.954962+00:00"},{"alias_kind":"arxiv_version","alias_value":"2605.21300v1","created_at":"2026-05-21T02:05:27.954962+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.21300","created_at":"2026-05-21T02:05:27.954962+00:00"},{"alias_kind":"pith_short_12","alias_value":"RPZKZWRIX4KJ","created_at":"2026-05-21T02:05:27.954962+00:00"},{"alias_kind":"pith_short_16","alias_value":"RPZKZWRIX4KJ4EVA","created_at":"2026-05-21T02:05:27.954962+00:00"},{"alias_kind":"pith_short_8","alias_value":"RPZKZWRI","created_at":"2026-05-21T02:05:27.954962+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/RPZKZWRIX4KJ4EVAYHN4WW5RHO","json":"https://pith.science/pith/RPZKZWRIX4KJ4EVAYHN4WW5RHO.json","graph_json":"https://pith.science/api/pith-number/RPZKZWRIX4KJ4EVAYHN4WW5RHO/graph.json","events_json":"https://pith.science/api/pith-number/RPZKZWRIX4KJ4EVAYHN4WW5RHO/events.json","paper":"https://pith.science/paper/RPZKZWRI"},"agent_actions":{"view_html":"https://pith.science/pith/RPZKZWRIX4KJ4EVAYHN4WW5RHO","download_json":"https://pith.science/pith/RPZKZWRIX4KJ4EVAYHN4WW5RHO.json","view_paper":"https://pith.science/paper/RPZKZWRI","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2605.21300&json=true","fetch_graph":"https://pith.science/api/pith-number/RPZKZWRIX4KJ4EVAYHN4WW5RHO/graph.json","fetch_events":"https://pith.science/api/pith-number/RPZKZWRIX4KJ4EVAYHN4WW5RHO/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/RPZKZWRIX4KJ4EVAYHN4WW5RHO/action/timestamp_anchor","attest_storage":"https://pith.science/pith/RPZKZWRIX4KJ4EVAYHN4WW5RHO/action/storage_attestation","attest_author":"https://pith.science/pith/RPZKZWRIX4KJ4EVAYHN4WW5RHO/action/author_attestation","sign_citation":"https://pith.science/pith/RPZKZWRIX4KJ4EVAYHN4WW5RHO/action/citation_signature","submit_replication":"https://pith.science/pith/RPZKZWRIX4KJ4EVAYHN4WW5RHO/action/replication_record"}},"created_at":"2026-05-21T02:05:27.954962+00:00","updated_at":"2026-05-21T02:05:27.954962+00:00"}