{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2022:SE4KIPDEAMK65QMQHTFRKHV2RU","short_pith_number":"pith:SE4KIPDE","schema_version":"1.0","canonical_sha256":"9138a43c640315eec1903ccb151eba8d31fcc838cd44bece75b0475c20faa39d","source":{"kind":"arxiv","id":"2207.00056","version":3},"attestation_state":"computed","paper":{"title":"MultiViz: Towards Visualizing and Understanding Multimodal Models","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.CL","cs.CV","cs.MM"],"primary_cat":"cs.LG","authors_text":"Gunjan Chhablani, Louis-Philippe Morency, Nihal Jain, Paul Pu Liang, Ruslan Salakhutdinov, Xingbo Wang, Yiwei Lyu, Zihao Deng","submitted_at":"2022-06-30T18:42:06Z","abstract_excerpt":"The promise of multimodal models for real-world applications has inspired research in visualizing and understanding their internal mechanics with the end goal of empowering stakeholders to visualize model behavior, perform model debugging, and promote trust in machine learning models. However, modern multimodal models are typically black-box neural networks, which makes it challenging to understand their internal mechanics. How can we visualize the internal modeling of multimodal interactions in these models? Our paper aims to fill this gap by proposing MultiViz, a method for analyzing the beh"},"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":"2207.00056","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2022-06-30T18:42:06Z","cross_cats_sorted":["cs.AI","cs.CL","cs.CV","cs.MM"],"title_canon_sha256":"df50882acd3231b7e25b883c20f339fa9712c51423547d03f78f42c91772273d","abstract_canon_sha256":"4167bc0d82a4dfedf8b404e940713c37433e969eb0b8abc3866702452e84f94a"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T05:48:34.844047Z","signature_b64":"4cKtNwySt316ieS7iipAlVnY0FkIs+QMWwpQF3f1quSgSG+akuIEHSkTgwuGo9cAOxCHFBsaM+EW0WEzYDuhBg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"9138a43c640315eec1903ccb151eba8d31fcc838cd44bece75b0475c20faa39d","last_reissued_at":"2026-07-05T05:48:34.843537Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T05:48:34.843537Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"MultiViz: Towards Visualizing and Understanding Multimodal Models","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.CL","cs.CV","cs.MM"],"primary_cat":"cs.LG","authors_text":"Gunjan Chhablani, Louis-Philippe Morency, Nihal Jain, Paul Pu Liang, Ruslan Salakhutdinov, Xingbo Wang, Yiwei Lyu, Zihao Deng","submitted_at":"2022-06-30T18:42:06Z","abstract_excerpt":"The promise of multimodal models for real-world applications has inspired research in visualizing and understanding their internal mechanics with the end goal of empowering stakeholders to visualize model behavior, perform model debugging, and promote trust in machine learning models. However, modern multimodal models are typically black-box neural networks, which makes it challenging to understand their internal mechanics. How can we visualize the internal modeling of multimodal interactions in these models? Our paper aims to fill this gap by proposing MultiViz, a method for analyzing the beh"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2207.00056","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/2207.00056/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":"2207.00056","created_at":"2026-07-05T05:48:34.843607+00:00"},{"alias_kind":"arxiv_version","alias_value":"2207.00056v3","created_at":"2026-07-05T05:48:34.843607+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2207.00056","created_at":"2026-07-05T05:48:34.843607+00:00"},{"alias_kind":"pith_short_12","alias_value":"SE4KIPDEAMK6","created_at":"2026-07-05T05:48:34.843607+00:00"},{"alias_kind":"pith_short_16","alias_value":"SE4KIPDEAMK65QMQ","created_at":"2026-07-05T05:48:34.843607+00:00"},{"alias_kind":"pith_short_8","alias_value":"SE4KIPDE","created_at":"2026-07-05T05:48:34.843607+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":2,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.09851","citing_title":"ECHO: Explainable Co-editing with Human-in-the-loop Operations for Presentation Refinement","ref_index":11,"is_internal_anchor":false},{"citing_arxiv_id":"2502.14888","citing_title":"Beyond Cross-Modal Alignment: Measuring and Leveraging Modality Gap in Vision-Language Models","ref_index":14,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/SE4KIPDEAMK65QMQHTFRKHV2RU","json":"https://pith.science/pith/SE4KIPDEAMK65QMQHTFRKHV2RU.json","graph_json":"https://pith.science/api/pith-number/SE4KIPDEAMK65QMQHTFRKHV2RU/graph.json","events_json":"https://pith.science/api/pith-number/SE4KIPDEAMK65QMQHTFRKHV2RU/events.json","paper":"https://pith.science/paper/SE4KIPDE"},"agent_actions":{"view_html":"https://pith.science/pith/SE4KIPDEAMK65QMQHTFRKHV2RU","download_json":"https://pith.science/pith/SE4KIPDEAMK65QMQHTFRKHV2RU.json","view_paper":"https://pith.science/paper/SE4KIPDE","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2207.00056&json=true","fetch_graph":"https://pith.science/api/pith-number/SE4KIPDEAMK65QMQHTFRKHV2RU/graph.json","fetch_events":"https://pith.science/api/pith-number/SE4KIPDEAMK65QMQHTFRKHV2RU/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/SE4KIPDEAMK65QMQHTFRKHV2RU/action/timestamp_anchor","attest_storage":"https://pith.science/pith/SE4KIPDEAMK65QMQHTFRKHV2RU/action/storage_attestation","attest_author":"https://pith.science/pith/SE4KIPDEAMK65QMQHTFRKHV2RU/action/author_attestation","sign_citation":"https://pith.science/pith/SE4KIPDEAMK65QMQHTFRKHV2RU/action/citation_signature","submit_replication":"https://pith.science/pith/SE4KIPDEAMK65QMQHTFRKHV2RU/action/replication_record"}},"created_at":"2026-07-05T05:48:34.843607+00:00","updated_at":"2026-07-05T05:48:34.843607+00:00"}