{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:7EOBPWI7MUUH7XNK5IFV75CGLU","short_pith_number":"pith:7EOBPWI7","schema_version":"1.0","canonical_sha256":"f91c17d91f65287fddaaea0b5ff4465d2019dd5d182c3a51fb3b606db43754ab","source":{"kind":"arxiv","id":"2606.04351","version":1},"attestation_state":"computed","paper":{"title":"Video2LoRA: Parametric Video Internalization for Vision-Language Models","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CL"],"primary_cat":"cs.CV","authors_text":"Dinesh Manocha, Manan Suri, Sarvesh Baskar","submitted_at":"2026-06-03T02:07:35Z","abstract_excerpt":"Processing video in vision-language models is expensive: each frame occupies hundreds of tokens, and inference cost scales with every frame and every repeated query. We introduce Video2LoRA, a method for parametric video internalization. A perceiver hypernetwork reads the intermediate representations produced layer-by-layer as a frozen VLM encodes a video, and generates a Low-Rank Adaptation (LoRA) adapter in a single forward pass. Unlike standard LoRA fine-tuning, which requires iterative gradient updates, Video2LoRA predicts these weights directly from the video. Trained for SmolVLM2 500M an"},"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":"2606.04351","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2026-06-03T02:07:35Z","cross_cats_sorted":["cs.CL"],"title_canon_sha256":"89b9d5de4e0ca7ad7e08e05a0d1d08c7d6e035ad58e2c58594604f29e364309a","abstract_canon_sha256":"e8ed70a9b16c719788b3403b90599f8c3d16f489f032dc3994ef1f37e656210a"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-04T01:09:04.533880Z","signature_b64":"ss2FqIrpLfYvYdtSImkSwFWn53qeKibD51/RpA5Q40eqVRAsnCvBmnFPmEQ8J+MW+c2bR2eeHWBSpeJcwc0JAA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"f91c17d91f65287fddaaea0b5ff4465d2019dd5d182c3a51fb3b606db43754ab","last_reissued_at":"2026-06-04T01:09:04.533340Z","signature_status":"signed_v1","first_computed_at":"2026-06-04T01:09:04.533340Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Video2LoRA: Parametric Video Internalization for Vision-Language Models","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CL"],"primary_cat":"cs.CV","authors_text":"Dinesh Manocha, Manan Suri, Sarvesh Baskar","submitted_at":"2026-06-03T02:07:35Z","abstract_excerpt":"Processing video in vision-language models is expensive: each frame occupies hundreds of tokens, and inference cost scales with every frame and every repeated query. We introduce Video2LoRA, a method for parametric video internalization. A perceiver hypernetwork reads the intermediate representations produced layer-by-layer as a frozen VLM encodes a video, and generates a Low-Rank Adaptation (LoRA) adapter in a single forward pass. Unlike standard LoRA fine-tuning, which requires iterative gradient updates, Video2LoRA predicts these weights directly from the video. Trained for SmolVLM2 500M an"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2606.04351","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/2606.04351/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":"2606.04351","created_at":"2026-06-04T01:09:04.533448+00:00"},{"alias_kind":"arxiv_version","alias_value":"2606.04351v1","created_at":"2026-06-04T01:09:04.533448+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2606.04351","created_at":"2026-06-04T01:09:04.533448+00:00"},{"alias_kind":"pith_short_12","alias_value":"7EOBPWI7MUUH","created_at":"2026-06-04T01:09:04.533448+00:00"},{"alias_kind":"pith_short_16","alias_value":"7EOBPWI7MUUH7XNK","created_at":"2026-06-04T01:09:04.533448+00:00"},{"alias_kind":"pith_short_8","alias_value":"7EOBPWI7","created_at":"2026-06-04T01:09:04.533448+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/7EOBPWI7MUUH7XNK5IFV75CGLU","json":"https://pith.science/pith/7EOBPWI7MUUH7XNK5IFV75CGLU.json","graph_json":"https://pith.science/api/pith-number/7EOBPWI7MUUH7XNK5IFV75CGLU/graph.json","events_json":"https://pith.science/api/pith-number/7EOBPWI7MUUH7XNK5IFV75CGLU/events.json","paper":"https://pith.science/paper/7EOBPWI7"},"agent_actions":{"view_html":"https://pith.science/pith/7EOBPWI7MUUH7XNK5IFV75CGLU","download_json":"https://pith.science/pith/7EOBPWI7MUUH7XNK5IFV75CGLU.json","view_paper":"https://pith.science/paper/7EOBPWI7","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2606.04351&json=true","fetch_graph":"https://pith.science/api/pith-number/7EOBPWI7MUUH7XNK5IFV75CGLU/graph.json","fetch_events":"https://pith.science/api/pith-number/7EOBPWI7MUUH7XNK5IFV75CGLU/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/7EOBPWI7MUUH7XNK5IFV75CGLU/action/timestamp_anchor","attest_storage":"https://pith.science/pith/7EOBPWI7MUUH7XNK5IFV75CGLU/action/storage_attestation","attest_author":"https://pith.science/pith/7EOBPWI7MUUH7XNK5IFV75CGLU/action/author_attestation","sign_citation":"https://pith.science/pith/7EOBPWI7MUUH7XNK5IFV75CGLU/action/citation_signature","submit_replication":"https://pith.science/pith/7EOBPWI7MUUH7XNK5IFV75CGLU/action/replication_record"}},"created_at":"2026-06-04T01:09:04.533448+00:00","updated_at":"2026-06-04T01:09:04.533448+00:00"}