{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:PF4CORSPRES5N6PYYWQLZBFWTP","short_pith_number":"pith:PF4CORSP","schema_version":"1.0","canonical_sha256":"797827464f8925d6f9f8c5a0bc84b69bf774bea450dba2311521167bba167cf0","source":{"kind":"arxiv","id":"2412.09582","version":2},"attestation_state":"computed","paper":{"title":"Neptune: The Long Orbit to Benchmarking Long Video Understanding","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.CV"],"primary_cat":"cs.LG","authors_text":"Arsha Nagrani, Austin Myers, Boqing Gong, Cordelia Schmid, Mikhail Sirotenko, Mingda Zhang, Nilpa Jha, Nitesh Bharadwaj Gundavarapu, Rachel Hornung, Ramin Mehran, Tobias Weyand, Xingyi Zhou, Yukun Zhu","submitted_at":"2024-12-12T18:54:48Z","abstract_excerpt":"We introduce Neptune, a benchmark for long video understanding that requires reasoning over long time horizons and across different modalities. Many existing video datasets and models are focused on short clips (10s-30s). While some long video datasets do exist, they can often be solved by powerful image models applied per frame (and often to very few frames) in a video, and are usually manually annotated at high cost. In order to mitigate both these problems, we propose a scalable dataset creation pipeline which leverages large models (VLMs and LLMs), to automatically generate dense, time-ali"},"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":"2412.09582","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2024-12-12T18:54:48Z","cross_cats_sorted":["cs.AI","cs.CV"],"title_canon_sha256":"6ac11784b7db87887baf202094038908db95c5114ee004f39c77bea0b621aae8","abstract_canon_sha256":"d19b1b9a2a75d7750cf4c8c3fcfb661c89d6b9ccfb41ada197e43cb12f98b40d"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:02:25.710508Z","signature_b64":"FPMb+pEGGeRd8XtUxy7vvPEWGkIldH0caQSXntwBAA6hEaGIgVjHzxCcmjorWb4BunDjG3XD0te5Gv6kzhunBA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"797827464f8925d6f9f8c5a0bc84b69bf774bea450dba2311521167bba167cf0","last_reissued_at":"2026-07-05T10:02:25.710046Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:02:25.710046Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Neptune: The Long Orbit to Benchmarking Long Video Understanding","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.CV"],"primary_cat":"cs.LG","authors_text":"Arsha Nagrani, Austin Myers, Boqing Gong, Cordelia Schmid, Mikhail Sirotenko, Mingda Zhang, Nilpa Jha, Nitesh Bharadwaj Gundavarapu, Rachel Hornung, Ramin Mehran, Tobias Weyand, Xingyi Zhou, Yukun Zhu","submitted_at":"2024-12-12T18:54:48Z","abstract_excerpt":"We introduce Neptune, a benchmark for long video understanding that requires reasoning over long time horizons and across different modalities. Many existing video datasets and models are focused on short clips (10s-30s). While some long video datasets do exist, they can often be solved by powerful image models applied per frame (and often to very few frames) in a video, and are usually manually annotated at high cost. In order to mitigate both these problems, we propose a scalable dataset creation pipeline which leverages large models (VLMs and LLMs), to automatically generate dense, time-ali"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2412.09582","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/2412.09582/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":"2412.09582","created_at":"2026-07-05T10:02:25.710103+00:00"},{"alias_kind":"arxiv_version","alias_value":"2412.09582v2","created_at":"2026-07-05T10:02:25.710103+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2412.09582","created_at":"2026-07-05T10:02:25.710103+00:00"},{"alias_kind":"pith_short_12","alias_value":"PF4CORSPRES5","created_at":"2026-07-05T10:02:25.710103+00:00"},{"alias_kind":"pith_short_16","alias_value":"PF4CORSPRES5N6PY","created_at":"2026-07-05T10:02:25.710103+00:00"},{"alias_kind":"pith_short_8","alias_value":"PF4CORSP","created_at":"2026-07-05T10:02:25.710103+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":4,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.20726","citing_title":"How Well Can Your Video Model Remember? Measuring Memory-Budget Trade-offs in Long Video Understanding","ref_index":15,"is_internal_anchor":false},{"citing_arxiv_id":"2601.06943","citing_title":"Watching, Reasoning, and Searching: A Video Deep Research Benchmark on Open Web for Agentic Video Reasoning","ref_index":10,"is_internal_anchor":false},{"citing_arxiv_id":"2605.15342","citing_title":"Minerva-Ego: Spatiotemporal Hints for Egocentric Video Understanding","ref_index":29,"is_internal_anchor":false},{"citing_arxiv_id":"2507.06261","citing_title":"Gemini 2.5: Pushing the Frontier with Advanced Reasoning, Multimodality, Long Context, and Next Generation Agentic Capabilities","ref_index":57,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/PF4CORSPRES5N6PYYWQLZBFWTP","json":"https://pith.science/pith/PF4CORSPRES5N6PYYWQLZBFWTP.json","graph_json":"https://pith.science/api/pith-number/PF4CORSPRES5N6PYYWQLZBFWTP/graph.json","events_json":"https://pith.science/api/pith-number/PF4CORSPRES5N6PYYWQLZBFWTP/events.json","paper":"https://pith.science/paper/PF4CORSP"},"agent_actions":{"view_html":"https://pith.science/pith/PF4CORSPRES5N6PYYWQLZBFWTP","download_json":"https://pith.science/pith/PF4CORSPRES5N6PYYWQLZBFWTP.json","view_paper":"https://pith.science/paper/PF4CORSP","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2412.09582&json=true","fetch_graph":"https://pith.science/api/pith-number/PF4CORSPRES5N6PYYWQLZBFWTP/graph.json","fetch_events":"https://pith.science/api/pith-number/PF4CORSPRES5N6PYYWQLZBFWTP/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/PF4CORSPRES5N6PYYWQLZBFWTP/action/timestamp_anchor","attest_storage":"https://pith.science/pith/PF4CORSPRES5N6PYYWQLZBFWTP/action/storage_attestation","attest_author":"https://pith.science/pith/PF4CORSPRES5N6PYYWQLZBFWTP/action/author_attestation","sign_citation":"https://pith.science/pith/PF4CORSPRES5N6PYYWQLZBFWTP/action/citation_signature","submit_replication":"https://pith.science/pith/PF4CORSPRES5N6PYYWQLZBFWTP/action/replication_record"}},"created_at":"2026-07-05T10:02:25.710103+00:00","updated_at":"2026-07-05T10:02:25.710103+00:00"}