{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:Y3QSIMF3KAOR5FL2YKN3F2Z27G","short_pith_number":"pith:Y3QSIMF3","schema_version":"1.0","canonical_sha256":"c6e12430bb501d1e957ac29bb2eb3af9a0d7d665be7622b50bdfb08351925464","source":{"kind":"arxiv","id":"2606.00232","version":1},"attestation_state":"computed","paper":{"title":"TIGER: Traceable Inference with Graph-Based Evidence Routing for Mitigating Hallucinations in Multimodal Generation","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.AI","authors_text":"Amanda Hughes, Kaixiang Zhao, Porter Jenkins, Shawn Huang, Tianrun Yu, Yushun Dong","submitted_at":"2026-05-29T18:06:26Z","abstract_excerpt":"We study fact-level repair for multimodal generation, where a fluent output may contain specific facts that are not supported by the input. Existing inference-time repair methods often generate feedback by jointly conditioning on the input and the current output. This design has two limitations: hallucinated claims in the output can bias the model's interpretation of the input, and free-form feedback cannot be ranked or scheduled at the fact level. We present TIGER, an inference-time framework that redesigns feedback for localized repair. TIGER independently extracts an observation graph from "},"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":"2606.00232","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2026-05-29T18:06:26Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"ee683ad1273bf6fbff9be2c21deff8a3f5e8ec78e6fc3d13f0734a4744b9556d","abstract_canon_sha256":"92d90c3c29ca1f939329648a054ef789372db0e77925b13eca15932ba58fffc9"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-02T01:03:22.347815Z","signature_b64":"eZ1A1k1wpW4UvMeo7AmRgPGKW0F0rQZn5yRM/t82b7gG7umt5pedStTqXbLrijXxIFpOf7eTOn9EFIRcyEdNBQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"c6e12430bb501d1e957ac29bb2eb3af9a0d7d665be7622b50bdfb08351925464","last_reissued_at":"2026-06-02T01:03:22.347453Z","signature_status":"signed_v1","first_computed_at":"2026-06-02T01:03:22.347453Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"TIGER: Traceable Inference with Graph-Based Evidence Routing for Mitigating Hallucinations in Multimodal Generation","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.AI","authors_text":"Amanda Hughes, Kaixiang Zhao, Porter Jenkins, Shawn Huang, Tianrun Yu, Yushun Dong","submitted_at":"2026-05-29T18:06:26Z","abstract_excerpt":"We study fact-level repair for multimodal generation, where a fluent output may contain specific facts that are not supported by the input. Existing inference-time repair methods often generate feedback by jointly conditioning on the input and the current output. This design has two limitations: hallucinated claims in the output can bias the model's interpretation of the input, and free-form feedback cannot be ranked or scheduled at the fact level. We present TIGER, an inference-time framework that redesigns feedback for localized repair. TIGER independently extracts an observation graph from "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2606.00232","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/2606.00232/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":"2606.00232","created_at":"2026-06-02T01:03:22.347506+00:00"},{"alias_kind":"arxiv_version","alias_value":"2606.00232v1","created_at":"2026-06-02T01:03:22.347506+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2606.00232","created_at":"2026-06-02T01:03:22.347506+00:00"},{"alias_kind":"pith_short_12","alias_value":"Y3QSIMF3KAOR","created_at":"2026-06-02T01:03:22.347506+00:00"},{"alias_kind":"pith_short_16","alias_value":"Y3QSIMF3KAOR5FL2","created_at":"2026-06-02T01:03:22.347506+00:00"},{"alias_kind":"pith_short_8","alias_value":"Y3QSIMF3","created_at":"2026-06-02T01:03:22.347506+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/Y3QSIMF3KAOR5FL2YKN3F2Z27G","json":"https://pith.science/pith/Y3QSIMF3KAOR5FL2YKN3F2Z27G.json","graph_json":"https://pith.science/api/pith-number/Y3QSIMF3KAOR5FL2YKN3F2Z27G/graph.json","events_json":"https://pith.science/api/pith-number/Y3QSIMF3KAOR5FL2YKN3F2Z27G/events.json","paper":"https://pith.science/paper/Y3QSIMF3"},"agent_actions":{"view_html":"https://pith.science/pith/Y3QSIMF3KAOR5FL2YKN3F2Z27G","download_json":"https://pith.science/pith/Y3QSIMF3KAOR5FL2YKN3F2Z27G.json","view_paper":"https://pith.science/paper/Y3QSIMF3","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2606.00232&json=true","fetch_graph":"https://pith.science/api/pith-number/Y3QSIMF3KAOR5FL2YKN3F2Z27G/graph.json","fetch_events":"https://pith.science/api/pith-number/Y3QSIMF3KAOR5FL2YKN3F2Z27G/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/Y3QSIMF3KAOR5FL2YKN3F2Z27G/action/timestamp_anchor","attest_storage":"https://pith.science/pith/Y3QSIMF3KAOR5FL2YKN3F2Z27G/action/storage_attestation","attest_author":"https://pith.science/pith/Y3QSIMF3KAOR5FL2YKN3F2Z27G/action/author_attestation","sign_citation":"https://pith.science/pith/Y3QSIMF3KAOR5FL2YKN3F2Z27G/action/citation_signature","submit_replication":"https://pith.science/pith/Y3QSIMF3KAOR5FL2YKN3F2Z27G/action/replication_record"}},"created_at":"2026-06-02T01:03:22.347506+00:00","updated_at":"2026-06-02T01:03:22.347506+00:00"}