{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2024:Q7ENI4Q5JRIYF3N3TDRGU7JOYM","short_pith_number":"pith:Q7ENI4Q5","canonical_record":{"source":{"id":"2405.20797","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2024-05-31T13:59:18Z","cross_cats_sorted":["cs.AI","cs.CL","cs.LG"],"title_canon_sha256":"21f76fe3800bb4bd41c7941b8fdc9003bc361dab331f0c2a35d222d3ab4f081b","abstract_canon_sha256":"a4d94a4ba7dbac9fa5edd4970ce241f9a49811568e133384ee3893496ee54c01"},"schema_version":"1.0"},"canonical_sha256":"87c8d4721d4c5182edbb98e26a7d2ec30678654ee6ab24d39257b8807adc3299","source":{"kind":"arxiv","id":"2405.20797","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2405.20797","created_at":"2026-07-05T08:32:41Z"},{"alias_kind":"arxiv_version","alias_value":"2405.20797v2","created_at":"2026-07-05T08:32:41Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2405.20797","created_at":"2026-07-05T08:32:41Z"},{"alias_kind":"pith_short_12","alias_value":"Q7ENI4Q5JRIY","created_at":"2026-07-05T08:32:41Z"},{"alias_kind":"pith_short_16","alias_value":"Q7ENI4Q5JRIYF3N3","created_at":"2026-07-05T08:32:41Z"},{"alias_kind":"pith_short_8","alias_value":"Q7ENI4Q5","created_at":"2026-07-05T08:32:41Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2024:Q7ENI4Q5JRIYF3N3TDRGU7JOYM","target":"record","payload":{"canonical_record":{"source":{"id":"2405.20797","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2024-05-31T13:59:18Z","cross_cats_sorted":["cs.AI","cs.CL","cs.LG"],"title_canon_sha256":"21f76fe3800bb4bd41c7941b8fdc9003bc361dab331f0c2a35d222d3ab4f081b","abstract_canon_sha256":"a4d94a4ba7dbac9fa5edd4970ce241f9a49811568e133384ee3893496ee54c01"},"schema_version":"1.0"},"canonical_sha256":"87c8d4721d4c5182edbb98e26a7d2ec30678654ee6ab24d39257b8807adc3299","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T08:32:41.519763Z","signature_b64":"nFUk37hta25uUcI6gfZ3Ok7crcfkzpPPDD52ygOe+QLmBb/7GYNXm6g8T6DnFJWu8VAqnNuQfOVvMD2c8aPcCQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"87c8d4721d4c5182edbb98e26a7d2ec30678654ee6ab24d39257b8807adc3299","last_reissued_at":"2026-07-05T08:32:41.519287Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T08:32:41.519287Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2405.20797","source_version":2,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T08:32:41Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"4ABy9Yw4AR2lEdbuOMkR0/IPhWkof0k/oGBphrcwrWQ+GmYI0ifEOHXv4Gyzv18k8tmnd6WEJ/6vLQRl7UwBCw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-04T04:22:22.927124Z"},"content_sha256":"454bbf8ebfc78cab0a719b772b2410af4b08fe275901ace08cc31dde72f94c38","schema_version":"1.0","event_id":"sha256:454bbf8ebfc78cab0a719b772b2410af4b08fe275901ace08cc31dde72f94c38"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2024:Q7ENI4Q5JRIYF3N3TDRGU7JOYM","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Ovis: Structural Embedding Alignment for Multimodal Large Language Model","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.CL","cs.LG"],"primary_cat":"cs.CV","authors_text":"Han-Jia Ye, Kaifu Zhang, Qing-Guo Chen, Shiyin Lu, Weihua Luo, Yang Li, Zhao Xu","submitted_at":"2024-05-31T13:59:18Z","abstract_excerpt":"Current Multimodal Large Language Models (MLLMs) typically integrate a pre-trained LLM with another pre-trained vision transformer through a connector, such as an MLP, endowing the LLM with visual capabilities. However, the misalignment between two embedding strategies in MLLMs -- the structural textual embeddings based on an embedding look-up table and the continuous embeddings generated directly by the vision encoder -- makes challenges for a more seamless fusion of visual and textual information. We propose Ovis, a novel MLLM architecture designed to structurally align visual and textual em"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2405.20797","kind":"arxiv","version":2},"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/2405.20797/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":"2026-07-05T08:32:41Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"dtjWDrIehLj4P72iSccQhWpE/P28rr062XbUByjT/GXqZpCuP0xDJCI0mZJmvGWtzrU5akm8Uxmev91A/sWSCA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-04T04:22:22.927653Z"},"content_sha256":"96a65942690090583c40dae15d4210b4d8d0f1210522ee53ea0d56248e76047b","schema_version":"1.0","event_id":"sha256:96a65942690090583c40dae15d4210b4d8d0f1210522ee53ea0d56248e76047b"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/Q7ENI4Q5JRIYF3N3TDRGU7JOYM/bundle.json","state_url":"https://pith.science/pith/Q7ENI4Q5JRIYF3N3TDRGU7JOYM/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/Q7ENI4Q5JRIYF3N3TDRGU7JOYM/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":"2026-08-04T04:22:22Z","links":{"resolver":"https://pith.science/pith/Q7ENI4Q5JRIYF3N3TDRGU7JOYM","bundle":"https://pith.science/pith/Q7ENI4Q5JRIYF3N3TDRGU7JOYM/bundle.json","state":"https://pith.science/pith/Q7ENI4Q5JRIYF3N3TDRGU7JOYM/state.json","well_known_bundle":"https://pith.science/.well-known/pith/Q7ENI4Q5JRIYF3N3TDRGU7JOYM/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:Q7ENI4Q5JRIYF3N3TDRGU7JOYM","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":"a4d94a4ba7dbac9fa5edd4970ce241f9a49811568e133384ee3893496ee54c01","cross_cats_sorted":["cs.AI","cs.CL","cs.LG"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2024-05-31T13:59:18Z","title_canon_sha256":"21f76fe3800bb4bd41c7941b8fdc9003bc361dab331f0c2a35d222d3ab4f081b"},"schema_version":"1.0","source":{"id":"2405.20797","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2405.20797","created_at":"2026-07-05T08:32:41Z"},{"alias_kind":"arxiv_version","alias_value":"2405.20797v2","created_at":"2026-07-05T08:32:41Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2405.20797","created_at":"2026-07-05T08:32:41Z"},{"alias_kind":"pith_short_12","alias_value":"Q7ENI4Q5JRIY","created_at":"2026-07-05T08:32:41Z"},{"alias_kind":"pith_short_16","alias_value":"Q7ENI4Q5JRIYF3N3","created_at":"2026-07-05T08:32:41Z"},{"alias_kind":"pith_short_8","alias_value":"Q7ENI4Q5","created_at":"2026-07-05T08:32:41Z"}],"graph_snapshots":[{"event_id":"sha256:96a65942690090583c40dae15d4210b4d8d0f1210522ee53ea0d56248e76047b","target":"graph","created_at":"2026-07-05T08:32:41Z","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/2405.20797/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Current Multimodal Large Language Models (MLLMs) typically integrate a pre-trained LLM with another pre-trained vision transformer through a connector, such as an MLP, endowing the LLM with visual capabilities. However, the misalignment between two embedding strategies in MLLMs -- the structural textual embeddings based on an embedding look-up table and the continuous embeddings generated directly by the vision encoder -- makes challenges for a more seamless fusion of visual and textual information. We propose Ovis, a novel MLLM architecture designed to structurally align visual and textual em","authors_text":"Han-Jia Ye, Kaifu Zhang, Qing-Guo Chen, Shiyin Lu, Weihua Luo, Yang Li, Zhao Xu","cross_cats":["cs.AI","cs.CL","cs.LG"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2024-05-31T13:59:18Z","title":"Ovis: Structural Embedding Alignment for Multimodal Large Language Model"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2405.20797","kind":"arxiv","version":2},"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":"sha256:454bbf8ebfc78cab0a719b772b2410af4b08fe275901ace08cc31dde72f94c38","target":"record","created_at":"2026-07-05T08:32:41Z","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":"a4d94a4ba7dbac9fa5edd4970ce241f9a49811568e133384ee3893496ee54c01","cross_cats_sorted":["cs.AI","cs.CL","cs.LG"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2024-05-31T13:59:18Z","title_canon_sha256":"21f76fe3800bb4bd41c7941b8fdc9003bc361dab331f0c2a35d222d3ab4f081b"},"schema_version":"1.0","source":{"id":"2405.20797","kind":"arxiv","version":2}},"canonical_sha256":"87c8d4721d4c5182edbb98e26a7d2ec30678654ee6ab24d39257b8807adc3299","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"87c8d4721d4c5182edbb98e26a7d2ec30678654ee6ab24d39257b8807adc3299","first_computed_at":"2026-07-05T08:32:41.519287Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T08:32:41.519287Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"nFUk37hta25uUcI6gfZ3Ok7crcfkzpPPDD52ygOe+QLmBb/7GYNXm6g8T6DnFJWu8VAqnNuQfOVvMD2c8aPcCQ==","signature_status":"signed_v1","signed_at":"2026-07-05T08:32:41.519763Z","signed_message":"canonical_sha256_bytes"},"source_id":"2405.20797","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:454bbf8ebfc78cab0a719b772b2410af4b08fe275901ace08cc31dde72f94c38","sha256:96a65942690090583c40dae15d4210b4d8d0f1210522ee53ea0d56248e76047b"],"state_sha256":"24921b1e30f7674059728b5cd9bc42769046487cae75a8f7f6da27c7d826bfcd"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"JWSnBB4P67i1x9bo4TxoLyQfr6JDurAb4a8eNUbkqqzH1AQqY2WnJo/BAZ5OXM081UYA9TVPWnoloJAJnmgFBw==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-04T04:22:22.931436Z","bundle_sha256":"012a16b2de56e8d874a575dc407550c1f430a80bd56c63fcb971b1e50eb67c64"}}