{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:N3FN3RB6VYYSH3BNJBVFRJ2MIY","short_pith_number":"pith:N3FN3RB6","schema_version":"1.0","canonical_sha256":"6ecaddc43eae3123ec2d486a58a74c4633442ccf9a9e2bade329de634b045169","source":{"kind":"arxiv","id":"2410.02763","version":1},"attestation_state":"computed","paper":{"title":"Vinoground: Scrutinizing LMMs over Dense Temporal Reasoning with Short Videos","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.CL","cs.LG"],"primary_cat":"cs.CV","authors_text":"Jianrui Zhang, Mu Cai, Yong Jae Lee","submitted_at":"2024-10-03T17:59:58Z","abstract_excerpt":"There has been growing sentiment recently that modern large multimodal models (LMMs) have addressed most of the key challenges related to short video comprehension. As a result, both academia and industry are gradually shifting their attention towards the more complex challenges posed by understanding long-form videos. However, is this really the case? Our studies indicate that LMMs still lack many fundamental reasoning capabilities even when dealing with short videos. We introduce Vinoground, a temporal counterfactual LMM evaluation benchmark encompassing 1000 short and natural video-caption "},"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":"2410.02763","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2024-10-03T17:59:58Z","cross_cats_sorted":["cs.AI","cs.CL","cs.LG"],"title_canon_sha256":"1a0b55aac4a84872fd1349d2c1c9070324ab3cf3e438e3931daaaa79565183b2","abstract_canon_sha256":"7e3b252f10aa523d3db37c14322e0debf8e86309a0fdbf26efe7a2893dbbf98f"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:15:24.832987Z","signature_b64":"Bz9UuS9mb9uLwDZnpJdkoyMKqufFV72jTtbzcWRFRuP4vxfZwROj+XhuHM8jCCAZrxVPTqwXawC0oq72EUQoBQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"6ecaddc43eae3123ec2d486a58a74c4633442ccf9a9e2bade329de634b045169","last_reissued_at":"2026-07-05T09:15:24.832550Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:15:24.832550Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Vinoground: Scrutinizing LMMs over Dense Temporal Reasoning with Short Videos","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.CL","cs.LG"],"primary_cat":"cs.CV","authors_text":"Jianrui Zhang, Mu Cai, Yong Jae Lee","submitted_at":"2024-10-03T17:59:58Z","abstract_excerpt":"There has been growing sentiment recently that modern large multimodal models (LMMs) have addressed most of the key challenges related to short video comprehension. As a result, both academia and industry are gradually shifting their attention towards the more complex challenges posed by understanding long-form videos. However, is this really the case? Our studies indicate that LMMs still lack many fundamental reasoning capabilities even when dealing with short videos. We introduce Vinoground, a temporal counterfactual LMM evaluation benchmark encompassing 1000 short and natural video-caption "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2410.02763","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/2410.02763/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":"2410.02763","created_at":"2026-07-05T09:15:24.832601+00:00"},{"alias_kind":"arxiv_version","alias_value":"2410.02763v1","created_at":"2026-07-05T09:15:24.832601+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2410.02763","created_at":"2026-07-05T09:15:24.832601+00:00"},{"alias_kind":"pith_short_12","alias_value":"N3FN3RB6VYYS","created_at":"2026-07-05T09:15:24.832601+00:00"},{"alias_kind":"pith_short_16","alias_value":"N3FN3RB6VYYSH3BN","created_at":"2026-07-05T09:15:24.832601+00:00"},{"alias_kind":"pith_short_8","alias_value":"N3FN3RB6","created_at":"2026-07-05T09:15:24.832601+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":5,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.01591","citing_title":"TLG: Temporal-Logic Grounding for Video Question Answering via Source-Annotation Reconstruction and Category-Targeted Reasoning","ref_index":2,"is_internal_anchor":false},{"citing_arxiv_id":"2605.22823","citing_title":"Which Way Did It Move? Diagnosing and Overcoming Directional Motion Blindness in Video-LLMs","ref_index":77,"is_internal_anchor":false},{"citing_arxiv_id":"2507.21420","citing_title":"ReGATE: Learning Faster and Better with Fewer Tokens in MLLMs","ref_index":45,"is_internal_anchor":false},{"citing_arxiv_id":"2604.00799","citing_title":"Multimodal Language Models Cannot Spot Spatial Inconsistencies","ref_index":37,"is_internal_anchor":false},{"citing_arxiv_id":"2604.11399","citing_title":"Lost in Adaptation: Layer-Selective Recovery of Temporal Reasoning in Video-Language Models","ref_index":16,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/N3FN3RB6VYYSH3BNJBVFRJ2MIY","json":"https://pith.science/pith/N3FN3RB6VYYSH3BNJBVFRJ2MIY.json","graph_json":"https://pith.science/api/pith-number/N3FN3RB6VYYSH3BNJBVFRJ2MIY/graph.json","events_json":"https://pith.science/api/pith-number/N3FN3RB6VYYSH3BNJBVFRJ2MIY/events.json","paper":"https://pith.science/paper/N3FN3RB6"},"agent_actions":{"view_html":"https://pith.science/pith/N3FN3RB6VYYSH3BNJBVFRJ2MIY","download_json":"https://pith.science/pith/N3FN3RB6VYYSH3BNJBVFRJ2MIY.json","view_paper":"https://pith.science/paper/N3FN3RB6","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2410.02763&json=true","fetch_graph":"https://pith.science/api/pith-number/N3FN3RB6VYYSH3BNJBVFRJ2MIY/graph.json","fetch_events":"https://pith.science/api/pith-number/N3FN3RB6VYYSH3BNJBVFRJ2MIY/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/N3FN3RB6VYYSH3BNJBVFRJ2MIY/action/timestamp_anchor","attest_storage":"https://pith.science/pith/N3FN3RB6VYYSH3BNJBVFRJ2MIY/action/storage_attestation","attest_author":"https://pith.science/pith/N3FN3RB6VYYSH3BNJBVFRJ2MIY/action/author_attestation","sign_citation":"https://pith.science/pith/N3FN3RB6VYYSH3BNJBVFRJ2MIY/action/citation_signature","submit_replication":"https://pith.science/pith/N3FN3RB6VYYSH3BNJBVFRJ2MIY/action/replication_record"}},"created_at":"2026-07-05T09:15:24.832601+00:00","updated_at":"2026-07-05T09:15:24.832601+00:00"}