{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:FUSOKYNOIJIFBP4INZZOUTTPL3","short_pith_number":"pith:FUSOKYNO","schema_version":"1.0","canonical_sha256":"2d24e561ae425050bf886e72ea4e6f5ed163e23e81c3d6afb037640df5216c87","source":{"kind":"arxiv","id":"2407.04681","version":1},"attestation_state":"computed","paper":{"title":"Rethinking Visual Prompting for Multimodal Large Language Models with External Knowledge","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.CL","cs.LG"],"primary_cat":"cs.CV","authors_text":"Dongdong Chen, Lu Yuan, Philip Torr, Ronald Clark, Weijian Xu, Yuanze Lin, Yunsheng Li","submitted_at":"2024-07-05T17:43:30Z","abstract_excerpt":"In recent years, multimodal large language models (MLLMs) have made significant strides by training on vast high-quality image-text datasets, enabling them to generally understand images well. However, the inherent difficulty in explicitly conveying fine-grained or spatially dense information in text, such as masks, poses a challenge for MLLMs, limiting their ability to answer questions requiring an understanding of detailed or localized visual elements. Drawing inspiration from the Retrieval-Augmented Generation (RAG) concept, this paper proposes a new visual prompt approach to integrate fine"},"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":"2407.04681","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2024-07-05T17:43:30Z","cross_cats_sorted":["cs.AI","cs.CL","cs.LG"],"title_canon_sha256":"d728afab82fa9b56454ef60c8e9e50adfa41e8c785d618c1f07cef8425462909","abstract_canon_sha256":"ba103a78bb7cc81273eb35815ef12723d8e73c82f5f448d3838084bfb07b322f"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T08:40:37.262271Z","signature_b64":"9wXJMr3IxPEdhmaA0wFbFwQbuAteE3FEkY5ZIxjXRbX+eALJ4bVhT865Jl53tViWlcjACB84zK3R0V+S/lwVBg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"2d24e561ae425050bf886e72ea4e6f5ed163e23e81c3d6afb037640df5216c87","last_reissued_at":"2026-07-05T08:40:37.261831Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T08:40:37.261831Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Rethinking Visual Prompting for Multimodal Large Language Models with External Knowledge","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.CL","cs.LG"],"primary_cat":"cs.CV","authors_text":"Dongdong Chen, Lu Yuan, Philip Torr, Ronald Clark, Weijian Xu, Yuanze Lin, Yunsheng Li","submitted_at":"2024-07-05T17:43:30Z","abstract_excerpt":"In recent years, multimodal large language models (MLLMs) have made significant strides by training on vast high-quality image-text datasets, enabling them to generally understand images well. However, the inherent difficulty in explicitly conveying fine-grained or spatially dense information in text, such as masks, poses a challenge for MLLMs, limiting their ability to answer questions requiring an understanding of detailed or localized visual elements. Drawing inspiration from the Retrieval-Augmented Generation (RAG) concept, this paper proposes a new visual prompt approach to integrate fine"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2407.04681","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/2407.04681/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":"2407.04681","created_at":"2026-07-05T08:40:37.261885+00:00"},{"alias_kind":"arxiv_version","alias_value":"2407.04681v1","created_at":"2026-07-05T08:40:37.261885+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2407.04681","created_at":"2026-07-05T08:40:37.261885+00:00"},{"alias_kind":"pith_short_12","alias_value":"FUSOKYNOIJIF","created_at":"2026-07-05T08:40:37.261885+00:00"},{"alias_kind":"pith_short_16","alias_value":"FUSOKYNOIJIFBP4I","created_at":"2026-07-05T08:40:37.261885+00:00"},{"alias_kind":"pith_short_8","alias_value":"FUSOKYNO","created_at":"2026-07-05T08:40:37.261885+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2412.09612","citing_title":"Olympus: A Universal Task Router for Computer Vision Tasks","ref_index":44,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/FUSOKYNOIJIFBP4INZZOUTTPL3","json":"https://pith.science/pith/FUSOKYNOIJIFBP4INZZOUTTPL3.json","graph_json":"https://pith.science/api/pith-number/FUSOKYNOIJIFBP4INZZOUTTPL3/graph.json","events_json":"https://pith.science/api/pith-number/FUSOKYNOIJIFBP4INZZOUTTPL3/events.json","paper":"https://pith.science/paper/FUSOKYNO"},"agent_actions":{"view_html":"https://pith.science/pith/FUSOKYNOIJIFBP4INZZOUTTPL3","download_json":"https://pith.science/pith/FUSOKYNOIJIFBP4INZZOUTTPL3.json","view_paper":"https://pith.science/paper/FUSOKYNO","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2407.04681&json=true","fetch_graph":"https://pith.science/api/pith-number/FUSOKYNOIJIFBP4INZZOUTTPL3/graph.json","fetch_events":"https://pith.science/api/pith-number/FUSOKYNOIJIFBP4INZZOUTTPL3/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/FUSOKYNOIJIFBP4INZZOUTTPL3/action/timestamp_anchor","attest_storage":"https://pith.science/pith/FUSOKYNOIJIFBP4INZZOUTTPL3/action/storage_attestation","attest_author":"https://pith.science/pith/FUSOKYNOIJIFBP4INZZOUTTPL3/action/author_attestation","sign_citation":"https://pith.science/pith/FUSOKYNOIJIFBP4INZZOUTTPL3/action/citation_signature","submit_replication":"https://pith.science/pith/FUSOKYNOIJIFBP4INZZOUTTPL3/action/replication_record"}},"created_at":"2026-07-05T08:40:37.261885+00:00","updated_at":"2026-07-05T08:40:37.261885+00:00"}