{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:PZBKTLETVROZY26WVZBLBTQXJ7","short_pith_number":"pith:PZBKTLET","schema_version":"1.0","canonical_sha256":"7e42a9ac93ac5d9c6bd6ae42b0ce174fd794704969dd523abf97c848c53e1da1","source":{"kind":"arxiv","id":"2501.01426","version":2},"attestation_state":"computed","paper":{"title":"Unifying Specialized Visual Encoders for Video Language Models","license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","headline":"","cross_cats":["cs.CL","cs.LG"],"primary_cat":"cs.CV","authors_text":"Honglu Zhou, Jihoon Chung, Juan Carlos Niebles, Max Gonzalez Saez-Diez, Olga Russakovsky, Tyler Zhu","submitted_at":"2025-01-02T18:59:45Z","abstract_excerpt":"The recent advent of Large Language Models (LLMs) has ushered sophisticated reasoning capabilities into the realm of video through Video Large Language Models (VideoLLMs). However, VideoLLMs currently rely on a single vision encoder for all of their visual processing, which limits the amount and type of visual information that can be conveyed to the LLM. Our method, MERV, Multi-Encoder Representation of Videos, instead leverages multiple frozen visual encoders to create a unified representation of a video, providing the VideoLLM with a comprehensive set of specialized visual knowledge. Spatio-"},"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":"2501.01426","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","primary_cat":"cs.CV","submitted_at":"2025-01-02T18:59:45Z","cross_cats_sorted":["cs.CL","cs.LG"],"title_canon_sha256":"45a43dfd953ed3a3af93ff42ce371f2d028ba5c5b7f7341b7f7eefdbb8399099","abstract_canon_sha256":"2891400d88cdaae331ec7a0ef7321082c81f1a314cd15377b3a130b6cd98099c"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:21:40.050860Z","signature_b64":"n+RvVzSPp9SwmVrENW4nph2WXEYv8ybH+PtuxiUFqUSf57E4bwx7pvpyeGCUy77kLjObRBO56mNauyVsA8DkBA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"7e42a9ac93ac5d9c6bd6ae42b0ce174fd794704969dd523abf97c848c53e1da1","last_reissued_at":"2026-07-05T11:21:40.050303Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:21:40.050303Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Unifying Specialized Visual Encoders for Video Language Models","license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","headline":"","cross_cats":["cs.CL","cs.LG"],"primary_cat":"cs.CV","authors_text":"Honglu Zhou, Jihoon Chung, Juan Carlos Niebles, Max Gonzalez Saez-Diez, Olga Russakovsky, Tyler Zhu","submitted_at":"2025-01-02T18:59:45Z","abstract_excerpt":"The recent advent of Large Language Models (LLMs) has ushered sophisticated reasoning capabilities into the realm of video through Video Large Language Models (VideoLLMs). However, VideoLLMs currently rely on a single vision encoder for all of their visual processing, which limits the amount and type of visual information that can be conveyed to the LLM. Our method, MERV, Multi-Encoder Representation of Videos, instead leverages multiple frozen visual encoders to create a unified representation of a video, providing the VideoLLM with a comprehensive set of specialized visual knowledge. Spatio-"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2501.01426","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/2501.01426/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":"2501.01426","created_at":"2026-07-05T11:21:40.050374+00:00"},{"alias_kind":"arxiv_version","alias_value":"2501.01426v2","created_at":"2026-07-05T11:21:40.050374+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2501.01426","created_at":"2026-07-05T11:21:40.050374+00:00"},{"alias_kind":"pith_short_12","alias_value":"PZBKTLETVROZ","created_at":"2026-07-05T11:21:40.050374+00:00"},{"alias_kind":"pith_short_16","alias_value":"PZBKTLETVROZY26W","created_at":"2026-07-05T11:21:40.050374+00:00"},{"alias_kind":"pith_short_8","alias_value":"PZBKTLET","created_at":"2026-07-05T11:21:40.050374+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2509.03501","citing_title":"Strefer: Empowering Video LLMs with Space-Time Referring and Reasoning via Synthetic Instruction Data","ref_index":10,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/PZBKTLETVROZY26WVZBLBTQXJ7","json":"https://pith.science/pith/PZBKTLETVROZY26WVZBLBTQXJ7.json","graph_json":"https://pith.science/api/pith-number/PZBKTLETVROZY26WVZBLBTQXJ7/graph.json","events_json":"https://pith.science/api/pith-number/PZBKTLETVROZY26WVZBLBTQXJ7/events.json","paper":"https://pith.science/paper/PZBKTLET"},"agent_actions":{"view_html":"https://pith.science/pith/PZBKTLETVROZY26WVZBLBTQXJ7","download_json":"https://pith.science/pith/PZBKTLETVROZY26WVZBLBTQXJ7.json","view_paper":"https://pith.science/paper/PZBKTLET","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2501.01426&json=true","fetch_graph":"https://pith.science/api/pith-number/PZBKTLETVROZY26WVZBLBTQXJ7/graph.json","fetch_events":"https://pith.science/api/pith-number/PZBKTLETVROZY26WVZBLBTQXJ7/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/PZBKTLETVROZY26WVZBLBTQXJ7/action/timestamp_anchor","attest_storage":"https://pith.science/pith/PZBKTLETVROZY26WVZBLBTQXJ7/action/storage_attestation","attest_author":"https://pith.science/pith/PZBKTLETVROZY26WVZBLBTQXJ7/action/author_attestation","sign_citation":"https://pith.science/pith/PZBKTLETVROZY26WVZBLBTQXJ7/action/citation_signature","submit_replication":"https://pith.science/pith/PZBKTLETVROZY26WVZBLBTQXJ7/action/replication_record"}},"created_at":"2026-07-05T11:21:40.050374+00:00","updated_at":"2026-07-05T11:21:40.050374+00:00"}