{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:A2MAQC3OS5U5ZCDGAW3IJPVIH2","short_pith_number":"pith:A2MAQC3O","schema_version":"1.0","canonical_sha256":"0698080b6e9769dc886605b684bea83e8b1905ea8479191fb13581b4cf1e412d","source":{"kind":"arxiv","id":"2608.03112","version":1},"attestation_state":"computed","paper":{"title":"Adaptive Two-Stage Visual Token Pruning for Efficient Inference in Video-Language Models","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CV","authors_text":"Chi Zhang, Heba Aly, Hongda Mao, Paribesh Regmi, Qingshuang Chen, Yelin Kim","submitted_at":"2026-08-04T04:32:48Z","abstract_excerpt":"Vision-language models excel at image and video understanding but suffer from high inference latency due to the need to process thousands of tokens per image, limiting their deployment on resource-constrained edge devices and in real-time surveillance applications. This challenge is further amplified in video processing, where multiple frames must be analyzed simultaneously. Existing token reduction techniques are largely developed for single-image inputs and therefore fail to account for the temporal and inter-frame redundancies present in video sequences. In addition, these methods generally"},"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":"2608.03112","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2026-08-04T04:32:48Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"c7e614825a1263f75a27a8424dd358fdc3dc6188c4e561b3f007a072c4c8879e","abstract_canon_sha256":"78b1ceea0a3c2e2aab2765225322231e544ca7a5e0478ff50dab1a46d535b949"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-08-05T00:45:42.550536Z","signature_b64":"aZTQSa68yc5t2YWpzkXFlhUHtt0a5nIRw/jRlEfjpGsOYS4yLUj8atm7ZMzxnoO0IxMTau1jHeAcyaC8HzbMCw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"0698080b6e9769dc886605b684bea83e8b1905ea8479191fb13581b4cf1e412d","last_reissued_at":"2026-08-05T00:45:42.548130Z","signature_status":"signed_v1","first_computed_at":"2026-08-05T00:45:42.548130Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Adaptive Two-Stage Visual Token Pruning for Efficient Inference in Video-Language Models","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CV","authors_text":"Chi Zhang, Heba Aly, Hongda Mao, Paribesh Regmi, Qingshuang Chen, Yelin Kim","submitted_at":"2026-08-04T04:32:48Z","abstract_excerpt":"Vision-language models excel at image and video understanding but suffer from high inference latency due to the need to process thousands of tokens per image, limiting their deployment on resource-constrained edge devices and in real-time surveillance applications. This challenge is further amplified in video processing, where multiple frames must be analyzed simultaneously. Existing token reduction techniques are largely developed for single-image inputs and therefore fail to account for the temporal and inter-frame redundancies present in video sequences. In addition, these methods generally"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2608.03112","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/2608.03112/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":"2608.03112","created_at":"2026-08-05T00:45:42.548965+00:00"},{"alias_kind":"arxiv_version","alias_value":"2608.03112v1","created_at":"2026-08-05T00:45:42.548965+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2608.03112","created_at":"2026-08-05T00:45:42.548965+00:00"},{"alias_kind":"pith_short_12","alias_value":"A2MAQC3OS5U5","created_at":"2026-08-05T00:45:42.548965+00:00"},{"alias_kind":"pith_short_16","alias_value":"A2MAQC3OS5U5ZCDG","created_at":"2026-08-05T00:45:42.548965+00:00"},{"alias_kind":"pith_short_8","alias_value":"A2MAQC3O","created_at":"2026-08-05T00:45:42.548965+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/A2MAQC3OS5U5ZCDGAW3IJPVIH2","json":"https://pith.science/pith/A2MAQC3OS5U5ZCDGAW3IJPVIH2.json","graph_json":"https://pith.science/api/pith-number/A2MAQC3OS5U5ZCDGAW3IJPVIH2/graph.json","events_json":"https://pith.science/api/pith-number/A2MAQC3OS5U5ZCDGAW3IJPVIH2/events.json","paper":"https://pith.science/paper/A2MAQC3O"},"agent_actions":{"view_html":"https://pith.science/pith/A2MAQC3OS5U5ZCDGAW3IJPVIH2","download_json":"https://pith.science/pith/A2MAQC3OS5U5ZCDGAW3IJPVIH2.json","view_paper":"https://pith.science/paper/A2MAQC3O","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2608.03112&json=true","fetch_graph":"https://pith.science/api/pith-number/A2MAQC3OS5U5ZCDGAW3IJPVIH2/graph.json","fetch_events":"https://pith.science/api/pith-number/A2MAQC3OS5U5ZCDGAW3IJPVIH2/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/A2MAQC3OS5U5ZCDGAW3IJPVIH2/action/timestamp_anchor","attest_storage":"https://pith.science/pith/A2MAQC3OS5U5ZCDGAW3IJPVIH2/action/storage_attestation","attest_author":"https://pith.science/pith/A2MAQC3OS5U5ZCDGAW3IJPVIH2/action/author_attestation","sign_citation":"https://pith.science/pith/A2MAQC3OS5U5ZCDGAW3IJPVIH2/action/citation_signature","submit_replication":"https://pith.science/pith/A2MAQC3OS5U5ZCDGAW3IJPVIH2/action/replication_record"}},"created_at":"2026-08-05T00:45:42.548965+00:00","updated_at":"2026-08-05T00:45:42.548965+00:00"}