{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2025:R24PPKZ5T7XMD642ZIA7OMLVYY","short_pith_number":"pith:R24PPKZ5","canonical_record":{"source":{"id":"2506.17901","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2025-06-22T05:11:46Z","cross_cats_sorted":[],"title_canon_sha256":"0e82280c54cfd192c6aae9c52a83bc6cfaa7497adaa69293f69ee97a4f90aa91","abstract_canon_sha256":"a9ce105960cab0257ba50a2a1696155cc7a9bb181dfa678711f72e5af4a4df27"},"schema_version":"1.0"},"canonical_sha256":"8eb8f7ab3d9feec1fb9aca01f73175c60ef562b03e77b1bfe8c10edabfc2e360","source":{"kind":"arxiv","id":"2506.17901","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2506.17901","created_at":"2026-07-05T11:25:25Z"},{"alias_kind":"arxiv_version","alias_value":"2506.17901v1","created_at":"2026-07-05T11:25:25Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2506.17901","created_at":"2026-07-05T11:25:25Z"},{"alias_kind":"pith_short_12","alias_value":"R24PPKZ5T7XM","created_at":"2026-07-05T11:25:25Z"},{"alias_kind":"pith_short_16","alias_value":"R24PPKZ5T7XMD642","created_at":"2026-07-05T11:25:25Z"},{"alias_kind":"pith_short_8","alias_value":"R24PPKZ5","created_at":"2026-07-05T11:25:25Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2025:R24PPKZ5T7XMD642ZIA7OMLVYY","target":"record","payload":{"canonical_record":{"source":{"id":"2506.17901","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2025-06-22T05:11:46Z","cross_cats_sorted":[],"title_canon_sha256":"0e82280c54cfd192c6aae9c52a83bc6cfaa7497adaa69293f69ee97a4f90aa91","abstract_canon_sha256":"a9ce105960cab0257ba50a2a1696155cc7a9bb181dfa678711f72e5af4a4df27"},"schema_version":"1.0"},"canonical_sha256":"8eb8f7ab3d9feec1fb9aca01f73175c60ef562b03e77b1bfe8c10edabfc2e360","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:25:25.156995Z","signature_b64":"IRVLG2fUKpsmcH0OCoi7Bv3LYQYDk1cIPJcWquKrdujG2meq63EWIBsJ0SmhS0f5I45/EG0dQqBbRXuPJGoPBw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"8eb8f7ab3d9feec1fb9aca01f73175c60ef562b03e77b1bfe8c10edabfc2e360","last_reissued_at":"2026-07-05T11:25:25.156593Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:25:25.156593Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2506.17901","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":"2026-07-05T11:25:25Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"0QwmPb8Fge92wZdqBW7ScJsE8jUBDdhsFZ+nFm3I2qY/ijmXnh61eMLdKYZCm/l7mm0s7iWJImHEeNWZBbhuAQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-03T18:35:41.750936Z"},"content_sha256":"c0e973e2061a3e3c918f4b3adfd94bf79bd11ca72fd6b76185836bf38f795e99","schema_version":"1.0","event_id":"sha256:c0e973e2061a3e3c918f4b3adfd94bf79bd11ca72fd6b76185836bf38f795e99"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2025:R24PPKZ5T7XMD642ZIA7OMLVYY","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"PostAlign: Multimodal Grounding as a Corrective Lens for MLLMs","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Jian Wu, Jindong Gu, Philip Torr, Yang Zhang, Yixuan Wu","submitted_at":"2025-06-22T05:11:46Z","abstract_excerpt":"Multimodal Large Language Models (MLLMs) excel in vision-language tasks, such as image captioning and visual question answering. However, they often suffer from over-reliance on spurious correlations, primarily due to linguistic priors that distract the model from leveraging actual visual information. To address these issues, we introduce MMGrounded-PostAlign, a post-multimodal alignment framework designed to enhance the visual understanding capabilities and mitigate the hallucinations of MLLMs. Our framework incorporates a multimodal grounding module for both visual grounding, which identifie"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2506.17901","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/2506.17901/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":"2026-07-05T11:25:25Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"D0f8gJW+lm+t2lNZvOFMJ2/uEeZrquTwO5Qf8Q3DmGCU57kHjEFkScaKInX051eM+PbznrHQogjoT4zW+S7ACg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-03T18:35:41.751833Z"},"content_sha256":"7d3f4dd3e82f837cc37427146003f73714faa18c2b2fe79e2295ca81277c0ec8","schema_version":"1.0","event_id":"sha256:7d3f4dd3e82f837cc37427146003f73714faa18c2b2fe79e2295ca81277c0ec8"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/R24PPKZ5T7XMD642ZIA7OMLVYY/bundle.json","state_url":"https://pith.science/pith/R24PPKZ5T7XMD642ZIA7OMLVYY/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/R24PPKZ5T7XMD642ZIA7OMLVYY/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":"2026-08-03T18:35:41Z","links":{"resolver":"https://pith.science/pith/R24PPKZ5T7XMD642ZIA7OMLVYY","bundle":"https://pith.science/pith/R24PPKZ5T7XMD642ZIA7OMLVYY/bundle.json","state":"https://pith.science/pith/R24PPKZ5T7XMD642ZIA7OMLVYY/state.json","well_known_bundle":"https://pith.science/.well-known/pith/R24PPKZ5T7XMD642ZIA7OMLVYY/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:R24PPKZ5T7XMD642ZIA7OMLVYY","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":"a9ce105960cab0257ba50a2a1696155cc7a9bb181dfa678711f72e5af4a4df27","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2025-06-22T05:11:46Z","title_canon_sha256":"0e82280c54cfd192c6aae9c52a83bc6cfaa7497adaa69293f69ee97a4f90aa91"},"schema_version":"1.0","source":{"id":"2506.17901","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2506.17901","created_at":"2026-07-05T11:25:25Z"},{"alias_kind":"arxiv_version","alias_value":"2506.17901v1","created_at":"2026-07-05T11:25:25Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2506.17901","created_at":"2026-07-05T11:25:25Z"},{"alias_kind":"pith_short_12","alias_value":"R24PPKZ5T7XM","created_at":"2026-07-05T11:25:25Z"},{"alias_kind":"pith_short_16","alias_value":"R24PPKZ5T7XMD642","created_at":"2026-07-05T11:25:25Z"},{"alias_kind":"pith_short_8","alias_value":"R24PPKZ5","created_at":"2026-07-05T11:25:25Z"}],"graph_snapshots":[{"event_id":"sha256:7d3f4dd3e82f837cc37427146003f73714faa18c2b2fe79e2295ca81277c0ec8","target":"graph","created_at":"2026-07-05T11:25:25Z","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/2506.17901/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Multimodal Large Language Models (MLLMs) excel in vision-language tasks, such as image captioning and visual question answering. However, they often suffer from over-reliance on spurious correlations, primarily due to linguistic priors that distract the model from leveraging actual visual information. To address these issues, we introduce MMGrounded-PostAlign, a post-multimodal alignment framework designed to enhance the visual understanding capabilities and mitigate the hallucinations of MLLMs. Our framework incorporates a multimodal grounding module for both visual grounding, which identifie","authors_text":"Jian Wu, Jindong Gu, Philip Torr, Yang Zhang, Yixuan Wu","cross_cats":[],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2025-06-22T05:11:46Z","title":"PostAlign: Multimodal Grounding as a Corrective Lens for MLLMs"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2506.17901","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:c0e973e2061a3e3c918f4b3adfd94bf79bd11ca72fd6b76185836bf38f795e99","target":"record","created_at":"2026-07-05T11:25:25Z","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":"a9ce105960cab0257ba50a2a1696155cc7a9bb181dfa678711f72e5af4a4df27","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2025-06-22T05:11:46Z","title_canon_sha256":"0e82280c54cfd192c6aae9c52a83bc6cfaa7497adaa69293f69ee97a4f90aa91"},"schema_version":"1.0","source":{"id":"2506.17901","kind":"arxiv","version":1}},"canonical_sha256":"8eb8f7ab3d9feec1fb9aca01f73175c60ef562b03e77b1bfe8c10edabfc2e360","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"8eb8f7ab3d9feec1fb9aca01f73175c60ef562b03e77b1bfe8c10edabfc2e360","first_computed_at":"2026-07-05T11:25:25.156593Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T11:25:25.156593Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"IRVLG2fUKpsmcH0OCoi7Bv3LYQYDk1cIPJcWquKrdujG2meq63EWIBsJ0SmhS0f5I45/EG0dQqBbRXuPJGoPBw==","signature_status":"signed_v1","signed_at":"2026-07-05T11:25:25.156995Z","signed_message":"canonical_sha256_bytes"},"source_id":"2506.17901","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:c0e973e2061a3e3c918f4b3adfd94bf79bd11ca72fd6b76185836bf38f795e99","sha256:7d3f4dd3e82f837cc37427146003f73714faa18c2b2fe79e2295ca81277c0ec8"],"state_sha256":"bd6d96367bf715884d175387422edb992be2699581a597b52c8b24c4f5614242"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"yrqR5ouwzkYO9KjszKTXcJFe0tkcQ418iwS/MCi+4u98bq+SvWbSrnHddv5UpTMn9RcaIcX+0T5e4/X32t9DAA==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-03T18:35:41.756048Z","bundle_sha256":"84ed094eff0179fe8001b89efbd6c29e1085821353b3c5af88c8863c631793b7"}}