{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:KFDCFXZOB2AVPLAZ27EDWI33I2","short_pith_number":"pith:KFDCFXZO","schema_version":"1.0","canonical_sha256":"514622df2e0e8157ac19d7c83b237b4682fd18b21504af976ca5f6452c8584a4","source":{"kind":"arxiv","id":"2305.17455","version":4},"attestation_state":"computed","paper":{"title":"CrossGET: Cross-Guided Ensemble of Tokens for Accelerating Vision-Language Transformers","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CL"],"primary_cat":"cs.CV","authors_text":"Anyi Rao, Chaofan Tao, Chun Yuan, Dachuan Shi, Jiaqi Wang, Zhendong Yang","submitted_at":"2023-05-27T12:07:21Z","abstract_excerpt":"Recent vision-language models have achieved tremendous advances. However, their computational costs are also escalating dramatically, making model acceleration exceedingly critical. To pursue more efficient vision-language Transformers, this paper introduces Cross-Guided Ensemble of Tokens (CrossGET), a general acceleration framework for vision-language Transformers. This framework adaptively combines tokens in real-time during inference, significantly reducing computational costs while maintaining high performance. CrossGET features two primary innovations: 1) Cross-Guided Matching and Ensemb"},"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":"2305.17455","kind":"arxiv","version":4},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2023-05-27T12:07:21Z","cross_cats_sorted":["cs.CL"],"title_canon_sha256":"05526705fbe112beee9d6b823128c2091fb6e539a5aa189023e3d7392345ff90","abstract_canon_sha256":"8fc89dcf9bcd4f6680cbd924f5945156002f9a8106dd81cc82df592b8ee75a8c"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T08:31:40.113690Z","signature_b64":"wupf0tL8bRb85kzjNuowrsSNxJZmMYH2q3LvGJlEXRRYjeMv97OJfkSJNztxmVM+pt+BAStcMcoAgb0rg0eWCQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"514622df2e0e8157ac19d7c83b237b4682fd18b21504af976ca5f6452c8584a4","last_reissued_at":"2026-07-05T08:31:40.113184Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T08:31:40.113184Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"CrossGET: Cross-Guided Ensemble of Tokens for Accelerating Vision-Language Transformers","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CL"],"primary_cat":"cs.CV","authors_text":"Anyi Rao, Chaofan Tao, Chun Yuan, Dachuan Shi, Jiaqi Wang, Zhendong Yang","submitted_at":"2023-05-27T12:07:21Z","abstract_excerpt":"Recent vision-language models have achieved tremendous advances. However, their computational costs are also escalating dramatically, making model acceleration exceedingly critical. To pursue more efficient vision-language Transformers, this paper introduces Cross-Guided Ensemble of Tokens (CrossGET), a general acceleration framework for vision-language Transformers. This framework adaptively combines tokens in real-time during inference, significantly reducing computational costs while maintaining high performance. CrossGET features two primary innovations: 1) Cross-Guided Matching and Ensemb"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2305.17455","kind":"arxiv","version":4},"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/2305.17455/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":"2305.17455","created_at":"2026-07-05T08:31:40.113238+00:00"},{"alias_kind":"arxiv_version","alias_value":"2305.17455v4","created_at":"2026-07-05T08:31:40.113238+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2305.17455","created_at":"2026-07-05T08:31:40.113238+00:00"},{"alias_kind":"pith_short_12","alias_value":"KFDCFXZOB2AV","created_at":"2026-07-05T08:31:40.113238+00:00"},{"alias_kind":"pith_short_16","alias_value":"KFDCFXZOB2AVPLAZ","created_at":"2026-07-05T08:31:40.113238+00:00"},{"alias_kind":"pith_short_8","alias_value":"KFDCFXZO","created_at":"2026-07-05T08:31:40.113238+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":3,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.24165","citing_title":"Spectral Evolution-Guided Token Pruning in Multimodal Large Language Models","ref_index":54,"is_internal_anchor":false},{"citing_arxiv_id":"2606.24156","citing_title":"Accelerating Multimodal Large Language Models with Prior-Corrected Token Reduction","ref_index":54,"is_internal_anchor":false},{"citing_arxiv_id":"2604.02689","citing_title":"Efficient3D: A Unified Framework for Adaptive and Debiased Token Reduction in 3D MLLMs","ref_index":60,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/KFDCFXZOB2AVPLAZ27EDWI33I2","json":"https://pith.science/pith/KFDCFXZOB2AVPLAZ27EDWI33I2.json","graph_json":"https://pith.science/api/pith-number/KFDCFXZOB2AVPLAZ27EDWI33I2/graph.json","events_json":"https://pith.science/api/pith-number/KFDCFXZOB2AVPLAZ27EDWI33I2/events.json","paper":"https://pith.science/paper/KFDCFXZO"},"agent_actions":{"view_html":"https://pith.science/pith/KFDCFXZOB2AVPLAZ27EDWI33I2","download_json":"https://pith.science/pith/KFDCFXZOB2AVPLAZ27EDWI33I2.json","view_paper":"https://pith.science/paper/KFDCFXZO","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2305.17455&json=true","fetch_graph":"https://pith.science/api/pith-number/KFDCFXZOB2AVPLAZ27EDWI33I2/graph.json","fetch_events":"https://pith.science/api/pith-number/KFDCFXZOB2AVPLAZ27EDWI33I2/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/KFDCFXZOB2AVPLAZ27EDWI33I2/action/timestamp_anchor","attest_storage":"https://pith.science/pith/KFDCFXZOB2AVPLAZ27EDWI33I2/action/storage_attestation","attest_author":"https://pith.science/pith/KFDCFXZOB2AVPLAZ27EDWI33I2/action/author_attestation","sign_citation":"https://pith.science/pith/KFDCFXZOB2AVPLAZ27EDWI33I2/action/citation_signature","submit_replication":"https://pith.science/pith/KFDCFXZOB2AVPLAZ27EDWI33I2/action/replication_record"}},"created_at":"2026-07-05T08:31:40.113238+00:00","updated_at":"2026-07-05T08:31:40.113238+00:00"}