{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2025:YYKGEA33VZ67JA5YQBOJJM6C6X","short_pith_number":"pith:YYKGEA33","canonical_record":{"source":{"id":"2505.05626","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2025-05-08T20:04:27Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"5c87a9b6ae75fd652e3d61a3c5d9218a907d1b4cb6419eac561824789379f9f6","abstract_canon_sha256":"246a8b3ab70a9d9c4c6a57f0ea3969efa7c89c587a6617ee74480294f268d1f7"},"schema_version":"1.0"},"canonical_sha256":"c61462037bae7df483b8805c94b3c2f5de572dbb15e0bc56f95830fe79609490","source":{"kind":"arxiv","id":"2505.05626","version":3},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2505.05626","created_at":"2026-07-05T11:30:43Z"},{"alias_kind":"arxiv_version","alias_value":"2505.05626v3","created_at":"2026-07-05T11:30:43Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2505.05626","created_at":"2026-07-05T11:30:43Z"},{"alias_kind":"pith_short_12","alias_value":"YYKGEA33VZ67","created_at":"2026-07-05T11:30:43Z"},{"alias_kind":"pith_short_16","alias_value":"YYKGEA33VZ67JA5Y","created_at":"2026-07-05T11:30:43Z"},{"alias_kind":"pith_short_8","alias_value":"YYKGEA33","created_at":"2026-07-05T11:30:43Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2025:YYKGEA33VZ67JA5YQBOJJM6C6X","target":"record","payload":{"canonical_record":{"source":{"id":"2505.05626","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2025-05-08T20:04:27Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"5c87a9b6ae75fd652e3d61a3c5d9218a907d1b4cb6419eac561824789379f9f6","abstract_canon_sha256":"246a8b3ab70a9d9c4c6a57f0ea3969efa7c89c587a6617ee74480294f268d1f7"},"schema_version":"1.0"},"canonical_sha256":"c61462037bae7df483b8805c94b3c2f5de572dbb15e0bc56f95830fe79609490","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:30:43.999001Z","signature_b64":"b0X2pcuVA6bf9s108Gn6DjsLklpnY4nVUBk9VaT4U878xi37eEsmboJXpi+oFcKSImNhZm0lBtTjklCPQMqlDw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"c61462037bae7df483b8805c94b3c2f5de572dbb15e0bc56f95830fe79609490","last_reissued_at":"2026-07-05T11:30:43.998514Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:30:43.998514Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2505.05626","source_version":3,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T11:30:43Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"T+fbuy7oWan6H0f2Iz8kMWvVUZa17xZCIm+f2zwoR+6e4BQvOGfwZ2fm+zdGwbmPMA++BXVJo2DdT8imKAk+Dw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-16T18:16:01.075148Z"},"content_sha256":"f83286659fb83e9161ecd7f828982a4eb1cf75721b01e0edb411bb3afa534adf","schema_version":"1.0","event_id":"sha256:f83286659fb83e9161ecd7f828982a4eb1cf75721b01e0edb411bb3afa534adf"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2025:YYKGEA33VZ67JA5YQBOJJM6C6X","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Perceiving Beyond Language Priors: Enhancing Visual Comprehension and Attention in Multimodal Models","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CV","authors_text":"Aarti Ghatkesar, Ganesh Venkatesh","submitted_at":"2025-05-08T20:04:27Z","abstract_excerpt":"Achieving deep alignment between vision and language remains a central challenge for Multimodal Large Language Models (MLLMs). These models often fail to fully leverage visual input, defaulting to strong language priors. Our approach first provides insights into how MLLMs internally build visual understanding of image regions and then introduces techniques to amplify this capability. Specifically, we explore techniques designed both to deepen the model's understanding of visual content and to ensure that these visual insights actively guide language generation. We demonstrate the superior mult"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2505.05626","kind":"arxiv","version":3},"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/2505.05626/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":"2026-07-05T11:30:43Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"GzEX/KLzhDnPBU2Hb8J21/iacywDcwduEDf3nZgDenXljLnZSg3zhuLJCIBokSrmWhrO9wp6HyOw9JJy2P6xBQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-16T18:16:01.075757Z"},"content_sha256":"6a35d88bd57634fa965ce8a8621737f686ff45eae599acb268f27c8dc34abd33","schema_version":"1.0","event_id":"sha256:6a35d88bd57634fa965ce8a8621737f686ff45eae599acb268f27c8dc34abd33"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/YYKGEA33VZ67JA5YQBOJJM6C6X/bundle.json","state_url":"https://pith.science/pith/YYKGEA33VZ67JA5YQBOJJM6C6X/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/YYKGEA33VZ67JA5YQBOJJM6C6X/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":"2026-08-16T18:16:01Z","links":{"resolver":"https://pith.science/pith/YYKGEA33VZ67JA5YQBOJJM6C6X","bundle":"https://pith.science/pith/YYKGEA33VZ67JA5YQBOJJM6C6X/bundle.json","state":"https://pith.science/pith/YYKGEA33VZ67JA5YQBOJJM6C6X/state.json","well_known_bundle":"https://pith.science/.well-known/pith/YYKGEA33VZ67JA5YQBOJJM6C6X/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:YYKGEA33VZ67JA5YQBOJJM6C6X","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":"246a8b3ab70a9d9c4c6a57f0ea3969efa7c89c587a6617ee74480294f268d1f7","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2025-05-08T20:04:27Z","title_canon_sha256":"5c87a9b6ae75fd652e3d61a3c5d9218a907d1b4cb6419eac561824789379f9f6"},"schema_version":"1.0","source":{"id":"2505.05626","kind":"arxiv","version":3}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2505.05626","created_at":"2026-07-05T11:30:43Z"},{"alias_kind":"arxiv_version","alias_value":"2505.05626v3","created_at":"2026-07-05T11:30:43Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2505.05626","created_at":"2026-07-05T11:30:43Z"},{"alias_kind":"pith_short_12","alias_value":"YYKGEA33VZ67","created_at":"2026-07-05T11:30:43Z"},{"alias_kind":"pith_short_16","alias_value":"YYKGEA33VZ67JA5Y","created_at":"2026-07-05T11:30:43Z"},{"alias_kind":"pith_short_8","alias_value":"YYKGEA33","created_at":"2026-07-05T11:30:43Z"}],"graph_snapshots":[{"event_id":"sha256:6a35d88bd57634fa965ce8a8621737f686ff45eae599acb268f27c8dc34abd33","target":"graph","created_at":"2026-07-05T11:30:43Z","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/2505.05626/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Achieving deep alignment between vision and language remains a central challenge for Multimodal Large Language Models (MLLMs). These models often fail to fully leverage visual input, defaulting to strong language priors. Our approach first provides insights into how MLLMs internally build visual understanding of image regions and then introduces techniques to amplify this capability. Specifically, we explore techniques designed both to deepen the model's understanding of visual content and to ensure that these visual insights actively guide language generation. We demonstrate the superior mult","authors_text":"Aarti Ghatkesar, Ganesh Venkatesh","cross_cats":["cs.AI"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2025-05-08T20:04:27Z","title":"Perceiving Beyond Language Priors: Enhancing Visual Comprehension and Attention in Multimodal Models"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2505.05626","kind":"arxiv","version":3},"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:f83286659fb83e9161ecd7f828982a4eb1cf75721b01e0edb411bb3afa534adf","target":"record","created_at":"2026-07-05T11:30:43Z","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":"246a8b3ab70a9d9c4c6a57f0ea3969efa7c89c587a6617ee74480294f268d1f7","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2025-05-08T20:04:27Z","title_canon_sha256":"5c87a9b6ae75fd652e3d61a3c5d9218a907d1b4cb6419eac561824789379f9f6"},"schema_version":"1.0","source":{"id":"2505.05626","kind":"arxiv","version":3}},"canonical_sha256":"c61462037bae7df483b8805c94b3c2f5de572dbb15e0bc56f95830fe79609490","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"c61462037bae7df483b8805c94b3c2f5de572dbb15e0bc56f95830fe79609490","first_computed_at":"2026-07-05T11:30:43.998514Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T11:30:43.998514Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"b0X2pcuVA6bf9s108Gn6DjsLklpnY4nVUBk9VaT4U878xi37eEsmboJXpi+oFcKSImNhZm0lBtTjklCPQMqlDw==","signature_status":"signed_v1","signed_at":"2026-07-05T11:30:43.999001Z","signed_message":"canonical_sha256_bytes"},"source_id":"2505.05626","source_kind":"arxiv","source_version":3}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:f83286659fb83e9161ecd7f828982a4eb1cf75721b01e0edb411bb3afa534adf","sha256:6a35d88bd57634fa965ce8a8621737f686ff45eae599acb268f27c8dc34abd33"],"state_sha256":"f2f43099d7d4e4adea2d70d3dcffbdcad8251a012a426c7c916c3a11a1a0a29d"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"XbCo2j5PXMV+hvFyrED/0Jg2ifSq3YKryK7BjA5wATkdMEA27WWbKWHgvqsu9SAIbwhrWIdW7/U5tQwC2c7IDQ==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-16T18:16:01.080878Z","bundle_sha256":"646537e2762c9e3c648157e375b9cacda2633e268cf5d0e68a970ccaba6d2ad5"}}