{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2025:AMX3BAS5LGA2HVZ44QMORDTSYS","short_pith_number":"pith:AMX3BAS5","canonical_record":{"source":{"id":"2508.02419","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2025-08-04T13:40:59Z","cross_cats_sorted":["cs.CL"],"title_canon_sha256":"c354049c25ba581e455754cf686e24f0a91ebdad6cfe0d6fa7ed7f6768da6bb1","abstract_canon_sha256":"28065b647909bf471342963ccec1a7eb827a113936b45cbeeb23517c90c8d34d"},"schema_version":"1.0"},"canonical_sha256":"032fb0825d5981a3d73ce418e88e72c483b14a3a957f9724a9a74a06b17297cc","source":{"kind":"arxiv","id":"2508.02419","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2508.02419","created_at":"2026-07-05T11:48:14Z"},{"alias_kind":"arxiv_version","alias_value":"2508.02419v1","created_at":"2026-07-05T11:48:14Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2508.02419","created_at":"2026-07-05T11:48:14Z"},{"alias_kind":"pith_short_12","alias_value":"AMX3BAS5LGA2","created_at":"2026-07-05T11:48:14Z"},{"alias_kind":"pith_short_16","alias_value":"AMX3BAS5LGA2HVZ4","created_at":"2026-07-05T11:48:14Z"},{"alias_kind":"pith_short_8","alias_value":"AMX3BAS5","created_at":"2026-07-05T11:48:14Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2025:AMX3BAS5LGA2HVZ44QMORDTSYS","target":"record","payload":{"canonical_record":{"source":{"id":"2508.02419","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2025-08-04T13:40:59Z","cross_cats_sorted":["cs.CL"],"title_canon_sha256":"c354049c25ba581e455754cf686e24f0a91ebdad6cfe0d6fa7ed7f6768da6bb1","abstract_canon_sha256":"28065b647909bf471342963ccec1a7eb827a113936b45cbeeb23517c90c8d34d"},"schema_version":"1.0"},"canonical_sha256":"032fb0825d5981a3d73ce418e88e72c483b14a3a957f9724a9a74a06b17297cc","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:48:14.043018Z","signature_b64":"fg//DE3+jwSsabrtXis85zpfxIvMxcglj+TVr2aLN2S3jYYjlfI0SnUxgnIPw3DnbmhWc2dM2JsznVFcHJjMDg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"032fb0825d5981a3d73ce418e88e72c483b14a3a957f9724a9a74a06b17297cc","last_reissued_at":"2026-07-05T11:48:14.042476Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:48:14.042476Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2508.02419","source_version":1,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T11:48:14Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"bUKx0N80yPyZAOoTitSURcXnCVLIzlT4LjY5gcT6YXO1TLlDPOEzRia05BdFX2Eyk53fT65U2/+77icB/qcIDw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-06T08:43:04.921160Z"},"content_sha256":"01e9e1c69baa17c0b5a730595ff73f0aefff00a1c0903506008082bd99fb0f07","schema_version":"1.0","event_id":"sha256:01e9e1c69baa17c0b5a730595ff73f0aefff00a1c0903506008082bd99fb0f07"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2025:AMX3BAS5LGA2HVZ44QMORDTSYS","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Modality Bias in LVLMs: Analyzing and Mitigating Object Hallucination via Attention Lens","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CL"],"primary_cat":"cs.CV","authors_text":"Haohan Zheng, Zhenguo Zhang","submitted_at":"2025-08-04T13:40:59Z","abstract_excerpt":"Large vision-language models (LVLMs) have demonstrated remarkable multimodal comprehension and reasoning capabilities, but they still suffer from severe object hallucination. Previous studies primarily attribute the flaw to linguistic prior caused by the scale mismatch between visual encoders and large language models (LLMs) in LVLMs. Specifically, as current LVLMs are built upon LLMs, they tend to over-rely on textual prompts and internal knowledge of LLMs, generating descriptions inconsistent with visual cues. However, through an in-depth investigation of the hallucinated mechanisms, we empi"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2508.02419","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/2508.02419/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"},"verdict_id":null},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T11:48:14Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"6Iobiqehc6VUPzoVcPoGAwcvoQXATv4fhfA9Gqbb4L8QszAqzAdDt3ycJ8L4ub6LBAr7bl/w9rKhT3X6WvSoBA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-06T08:43:04.921558Z"},"content_sha256":"e7ce9a132633ef51b5dd8e15b1ec468ae2e2a5b67d2c902e7d8d820792a57986","schema_version":"1.0","event_id":"sha256:e7ce9a132633ef51b5dd8e15b1ec468ae2e2a5b67d2c902e7d8d820792a57986"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/AMX3BAS5LGA2HVZ44QMORDTSYS/bundle.json","state_url":"https://pith.science/pith/AMX3BAS5LGA2HVZ44QMORDTSYS/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/AMX3BAS5LGA2HVZ44QMORDTSYS/bundle.json","status":"primary"}],"public_keys":[{"key_id":"pith-v1-2026-05","algorithm":"ed25519","format":"raw","public_key_b64":"stVStoiQhXFxp4s2pdzPNoqVNBMojDU/fJ2db5S3CbM=","public_key_hex":"b2d552b68890857171a78b36a5dccf368a953413288c353f7c9d9d6f94b709b3","fingerprint_sha256_b32_first128bits":"RVFV5Z2OI2J3ZUO7ERDEBCYNKS","fingerprint_sha256_hex":"8d4b5ee74e4693bcd1df2446408b0d54","rotates_at":null,"url":"https://pith.science/pith-signing-key.json","notes":"Pith uses this Ed25519 key to sign canonical record SHA-256 digests. Verify with: ed25519_verify(public_key, message=canonical_sha256_bytes, signature=base64decode(signature_b64))."}],"merge_version":"pith-open-graph-merge-v1","built_at":"2026-07-06T08:43:04Z","links":{"resolver":"https://pith.science/pith/AMX3BAS5LGA2HVZ44QMORDTSYS","bundle":"https://pith.science/pith/AMX3BAS5LGA2HVZ44QMORDTSYS/bundle.json","state":"https://pith.science/pith/AMX3BAS5LGA2HVZ44QMORDTSYS/state.json","well_known_bundle":"https://pith.science/.well-known/pith/AMX3BAS5LGA2HVZ44QMORDTSYS/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:AMX3BAS5LGA2HVZ44QMORDTSYS","merge_version":"pith-open-graph-merge-v1","event_count":2,"valid_event_count":2,"invalid_event_count":0,"equivocation_count":0,"current":{"canonical_record":{"metadata":{"abstract_canon_sha256":"28065b647909bf471342963ccec1a7eb827a113936b45cbeeb23517c90c8d34d","cross_cats_sorted":["cs.CL"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2025-08-04T13:40:59Z","title_canon_sha256":"c354049c25ba581e455754cf686e24f0a91ebdad6cfe0d6fa7ed7f6768da6bb1"},"schema_version":"1.0","source":{"id":"2508.02419","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2508.02419","created_at":"2026-07-05T11:48:14Z"},{"alias_kind":"arxiv_version","alias_value":"2508.02419v1","created_at":"2026-07-05T11:48:14Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2508.02419","created_at":"2026-07-05T11:48:14Z"},{"alias_kind":"pith_short_12","alias_value":"AMX3BAS5LGA2","created_at":"2026-07-05T11:48:14Z"},{"alias_kind":"pith_short_16","alias_value":"AMX3BAS5LGA2HVZ4","created_at":"2026-07-05T11:48:14Z"},{"alias_kind":"pith_short_8","alias_value":"AMX3BAS5","created_at":"2026-07-05T11:48:14Z"}],"graph_snapshots":[{"event_id":"sha256:e7ce9a132633ef51b5dd8e15b1ec468ae2e2a5b67d2c902e7d8d820792a57986","target":"graph","created_at":"2026-07-05T11:48:14Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"graph_snapshot":{"author_claims":{"count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","strong_count":0},"builder_version":"pith-number-builder-2026-05-17-v1","claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"integrity":{"available":true,"clean":true,"detectors_run":[],"endpoint":"/pith/2508.02419/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Large vision-language models (LVLMs) have demonstrated remarkable multimodal comprehension and reasoning capabilities, but they still suffer from severe object hallucination. Previous studies primarily attribute the flaw to linguistic prior caused by the scale mismatch between visual encoders and large language models (LLMs) in LVLMs. Specifically, as current LVLMs are built upon LLMs, they tend to over-rely on textual prompts and internal knowledge of LLMs, generating descriptions inconsistent with visual cues. However, through an in-depth investigation of the hallucinated mechanisms, we empi","authors_text":"Haohan Zheng, Zhenguo Zhang","cross_cats":["cs.CL"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2025-08-04T13:40:59Z","title":"Modality Bias in LVLMs: Analyzing and Mitigating Object Hallucination via Attention Lens"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2508.02419","kind":"arxiv","version":1},"verdict":{"created_at":null,"id":null,"model_set":{},"one_line_summary":"","pipeline_version":null,"pith_extraction_headline":"","strongest_claim":"","weakest_assumption":""}},"verdict_id":null}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:01e9e1c69baa17c0b5a730595ff73f0aefff00a1c0903506008082bd99fb0f07","target":"record","created_at":"2026-07-05T11:48:14Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"attestation_state":"computed","canonical_record":{"metadata":{"abstract_canon_sha256":"28065b647909bf471342963ccec1a7eb827a113936b45cbeeb23517c90c8d34d","cross_cats_sorted":["cs.CL"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2025-08-04T13:40:59Z","title_canon_sha256":"c354049c25ba581e455754cf686e24f0a91ebdad6cfe0d6fa7ed7f6768da6bb1"},"schema_version":"1.0","source":{"id":"2508.02419","kind":"arxiv","version":1}},"canonical_sha256":"032fb0825d5981a3d73ce418e88e72c483b14a3a957f9724a9a74a06b17297cc","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"032fb0825d5981a3d73ce418e88e72c483b14a3a957f9724a9a74a06b17297cc","first_computed_at":"2026-07-05T11:48:14.042476Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T11:48:14.042476Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"fg//DE3+jwSsabrtXis85zpfxIvMxcglj+TVr2aLN2S3jYYjlfI0SnUxgnIPw3DnbmhWc2dM2JsznVFcHJjMDg==","signature_status":"signed_v1","signed_at":"2026-07-05T11:48:14.043018Z","signed_message":"canonical_sha256_bytes"},"source_id":"2508.02419","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:01e9e1c69baa17c0b5a730595ff73f0aefff00a1c0903506008082bd99fb0f07","sha256:e7ce9a132633ef51b5dd8e15b1ec468ae2e2a5b67d2c902e7d8d820792a57986"],"state_sha256":"bb9f894f966021c4ec0e34e93e2de0f5e4b9563addb361b80a6c792e100f3181"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"8vM0EIOJJsCZHwnhn1rnlPCC38tU95oo80iKyM5J3Kho/+BLvDNHWGpm8wLYR6VuSRV1BDfXgU6+8mIG2vqkAg==","signed_message":"bundle_sha256_bytes","signed_at":"2026-07-06T08:43:04.923495Z","bundle_sha256":"76cfce8cc22b851ef112f757303ac4aa8324549f092353602564b3cef17fc50e"}}