{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:2I2XDWUKIYTZ6HRT3OSQMLQZMI","short_pith_number":"pith:2I2XDWUK","schema_version":"1.0","canonical_sha256":"d23571da8a46279f1e33dba5062e19622d9f2471e947ea774452f09942477a3c","source":{"kind":"arxiv","id":"2412.12785","version":2},"attestation_state":"computed","paper":{"title":"Activating Distributed Visual Region within LLMs for Efficient and Effective Vision-Language Training and Inference","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Chengxing Zhou, Dianyi Wang, Siyuan Wang, Xuanjing Huang, Zejun Li, Zhihao Fan, Zhongyu Wei","submitted_at":"2024-12-17T10:44:47Z","abstract_excerpt":"Large Vision-Language Models (LVLMs) typically learn visual capacity through visual instruction tuning, involving updates to both a projector and their LLM backbones. Inspired by the concept of a visual region in the human brain, we investigate the existence of an analogous \\textit{visual region} within LLMs that functions as a cognitive core, and explore the potential of efficient training of LVLMs via selective layers tuning. Using Bunny-Llama-3-8B-V for detailed analysis and other three LVLMs for validation across diverse visual and textual tasks, we find that selectively updating 25\\% of L"},"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":"2412.12785","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2024-12-17T10:44:47Z","cross_cats_sorted":[],"title_canon_sha256":"ff3212dde8453d34659706988847ed90296c1197243a3859273ad090b936aa39","abstract_canon_sha256":"fb66ad8edf88e99f8c5faeca2946e911bf24e16c9f6ba1cb1cc112f86a4f43b9"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:36:33.206811Z","signature_b64":"XHttEfuAJEqpvATsPRWUpetlfRPzK030zrZBthoZ+6Mfk86hyruXfXTwcals69m2lYKeRV7z/hZPMMWHaGFDBg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"d23571da8a46279f1e33dba5062e19622d9f2471e947ea774452f09942477a3c","last_reissued_at":"2026-07-05T10:36:33.206128Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:36:33.206128Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Activating Distributed Visual Region within LLMs for Efficient and Effective Vision-Language Training and Inference","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Chengxing Zhou, Dianyi Wang, Siyuan Wang, Xuanjing Huang, Zejun Li, Zhihao Fan, Zhongyu Wei","submitted_at":"2024-12-17T10:44:47Z","abstract_excerpt":"Large Vision-Language Models (LVLMs) typically learn visual capacity through visual instruction tuning, involving updates to both a projector and their LLM backbones. Inspired by the concept of a visual region in the human brain, we investigate the existence of an analogous \\textit{visual region} within LLMs that functions as a cognitive core, and explore the potential of efficient training of LVLMs via selective layers tuning. Using Bunny-Llama-3-8B-V for detailed analysis and other three LVLMs for validation across diverse visual and textual tasks, we find that selectively updating 25\\% of L"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2412.12785","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/2412.12785/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":"2412.12785","created_at":"2026-07-05T10:36:33.206200+00:00"},{"alias_kind":"arxiv_version","alias_value":"2412.12785v2","created_at":"2026-07-05T10:36:33.206200+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2412.12785","created_at":"2026-07-05T10:36:33.206200+00:00"},{"alias_kind":"pith_short_12","alias_value":"2I2XDWUKIYTZ","created_at":"2026-07-05T10:36:33.206200+00:00"},{"alias_kind":"pith_short_16","alias_value":"2I2XDWUKIYTZ6HRT","created_at":"2026-07-05T10:36:33.206200+00:00"},{"alias_kind":"pith_short_8","alias_value":"2I2XDWUK","created_at":"2026-07-05T10:36:33.206200+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/2I2XDWUKIYTZ6HRT3OSQMLQZMI","json":"https://pith.science/pith/2I2XDWUKIYTZ6HRT3OSQMLQZMI.json","graph_json":"https://pith.science/api/pith-number/2I2XDWUKIYTZ6HRT3OSQMLQZMI/graph.json","events_json":"https://pith.science/api/pith-number/2I2XDWUKIYTZ6HRT3OSQMLQZMI/events.json","paper":"https://pith.science/paper/2I2XDWUK"},"agent_actions":{"view_html":"https://pith.science/pith/2I2XDWUKIYTZ6HRT3OSQMLQZMI","download_json":"https://pith.science/pith/2I2XDWUKIYTZ6HRT3OSQMLQZMI.json","view_paper":"https://pith.science/paper/2I2XDWUK","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2412.12785&json=true","fetch_graph":"https://pith.science/api/pith-number/2I2XDWUKIYTZ6HRT3OSQMLQZMI/graph.json","fetch_events":"https://pith.science/api/pith-number/2I2XDWUKIYTZ6HRT3OSQMLQZMI/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/2I2XDWUKIYTZ6HRT3OSQMLQZMI/action/timestamp_anchor","attest_storage":"https://pith.science/pith/2I2XDWUKIYTZ6HRT3OSQMLQZMI/action/storage_attestation","attest_author":"https://pith.science/pith/2I2XDWUKIYTZ6HRT3OSQMLQZMI/action/author_attestation","sign_citation":"https://pith.science/pith/2I2XDWUKIYTZ6HRT3OSQMLQZMI/action/citation_signature","submit_replication":"https://pith.science/pith/2I2XDWUKIYTZ6HRT3OSQMLQZMI/action/replication_record"}},"created_at":"2026-07-05T10:36:33.206200+00:00","updated_at":"2026-07-05T10:36:33.206200+00:00"}