{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:Y3SDCMFMWYSRI7NPUAA5X4LFT3","short_pith_number":"pith:Y3SDCMFM","schema_version":"1.0","canonical_sha256":"c6e43130acb625147dafa001dbf1659ed129a7f107dca852b72029a4e4026297","source":{"kind":"arxiv","id":"2404.07973","version":1},"attestation_state":"computed","paper":{"title":"Ferret-v2: An Improved Baseline for Referring and Grounding with Large Language Models","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Bowen Zhang, Chen Chen, Haotian Zhang, Haoxuan You, Hong-You Chen, Philipp Dufter, Shih-Fu Chang, Tsu-Jui Fu, William Yang Wang, Yinfei Yang, Zhe Gan","submitted_at":"2024-04-11T17:56:05Z","abstract_excerpt":"While Ferret seamlessly integrates regional understanding into the Large Language Model (LLM) to facilitate its referring and grounding capability, it poses certain limitations: constrained by the pre-trained fixed visual encoder and failed to perform well on broader tasks. In this work, we unveil Ferret-v2, a significant upgrade to Ferret, with three key designs. (1) Any resolution grounding and referring: A flexible approach that effortlessly handles higher image resolution, improving the model's ability to process and understand images in greater detail. (2) Multi-granularity visual encodin"},"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":"2404.07973","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2024-04-11T17:56:05Z","cross_cats_sorted":[],"title_canon_sha256":"495d1ecd635701c8271579eea72023f7dc9a8d3e1f310e0d6c4c9f01560ff633","abstract_canon_sha256":"64faee92267ddecd3ab855cc26741645f5a659fa0d2757bf14170900605daec5"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T08:07:00.068551Z","signature_b64":"5LV8iOkYvqqiMZlZEc5DTJrUfu41itS3Fo0vEigTjeTK9lHc/VRw0705yH8Tmsib7P7c30mfdhsDw9WXbix6DA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"c6e43130acb625147dafa001dbf1659ed129a7f107dca852b72029a4e4026297","last_reissued_at":"2026-07-05T08:07:00.068055Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T08:07:00.068055Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Ferret-v2: An Improved Baseline for Referring and Grounding with Large Language Models","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Bowen Zhang, Chen Chen, Haotian Zhang, Haoxuan You, Hong-You Chen, Philipp Dufter, Shih-Fu Chang, Tsu-Jui Fu, William Yang Wang, Yinfei Yang, Zhe Gan","submitted_at":"2024-04-11T17:56:05Z","abstract_excerpt":"While Ferret seamlessly integrates regional understanding into the Large Language Model (LLM) to facilitate its referring and grounding capability, it poses certain limitations: constrained by the pre-trained fixed visual encoder and failed to perform well on broader tasks. In this work, we unveil Ferret-v2, a significant upgrade to Ferret, with three key designs. (1) Any resolution grounding and referring: A flexible approach that effortlessly handles higher image resolution, improving the model's ability to process and understand images in greater detail. (2) Multi-granularity visual encodin"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2404.07973","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/2404.07973/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":"2404.07973","created_at":"2026-07-05T08:07:00.068119+00:00"},{"alias_kind":"arxiv_version","alias_value":"2404.07973v1","created_at":"2026-07-05T08:07:00.068119+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2404.07973","created_at":"2026-07-05T08:07:00.068119+00:00"},{"alias_kind":"pith_short_12","alias_value":"Y3SDCMFMWYSR","created_at":"2026-07-05T08:07:00.068119+00:00"},{"alias_kind":"pith_short_16","alias_value":"Y3SDCMFMWYSRI7NP","created_at":"2026-07-05T08:07:00.068119+00:00"},{"alias_kind":"pith_short_8","alias_value":"Y3SDCMFM","created_at":"2026-07-05T08:07:00.068119+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":14,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.17030","citing_title":"Qwen-RobotWorld Technical Report: Unifying Embodied World Modeling through Language-Conditioned Video Generation","ref_index":181,"is_internal_anchor":false},{"citing_arxiv_id":"2605.14923","citing_title":"SceneParser: Hierarchical Scene Parsing for Visual Semantics Understanding","ref_index":15,"is_internal_anchor":false},{"citing_arxiv_id":"2502.13923","citing_title":"Qwen2.5-VL Technical Report","ref_index":40,"is_internal_anchor":false},{"citing_arxiv_id":"2605.18018","citing_title":"See What I Mean: Aligning Vision and Language Representations for Video Fine-grained Object Understanding","ref_index":98,"is_internal_anchor":false},{"citing_arxiv_id":"2512.10554","citing_title":"Grounding Everything in Tokens for Multimodal Large Language Models","ref_index":87,"is_internal_anchor":false},{"citing_arxiv_id":"2605.11462","citing_title":"SpatialForge: Bootstrapping 3D-Aware Spatial Reasoning from Open-World 2D Images","ref_index":36,"is_internal_anchor":false},{"citing_arxiv_id":"2407.07726","citing_title":"PaliGemma: A versatile 3B VLM for transfer","ref_index":163,"is_internal_anchor":false},{"citing_arxiv_id":"2604.11789","citing_title":"LMMs Meet Object-Centric Vision: Understanding, Segmentation, Editing and Generation","ref_index":224,"is_internal_anchor":false},{"citing_arxiv_id":"2412.10302","citing_title":"DeepSeek-VL2: Mixture-of-Experts Vision-Language Models for Advanced Multimodal Understanding","ref_index":109,"is_internal_anchor":false},{"citing_arxiv_id":"2504.10479","citing_title":"InternVL3: Exploring Advanced Training and Test-Time Recipes for Open-Source Multimodal Models","ref_index":145,"is_internal_anchor":false},{"citing_arxiv_id":"2412.05271","citing_title":"Expanding Performance Boundaries of Open-Source Multimodal Models with Model, Data, and Test-Time Scaling","ref_index":298,"is_internal_anchor":false},{"citing_arxiv_id":"2508.18265","citing_title":"InternVL3.5: Advancing Open-Source Multimodal Models in Versatility, Reasoning, and Efficiency","ref_index":174,"is_internal_anchor":false},{"citing_arxiv_id":"2604.22836","citing_title":"AgentRVOS for MeViS-Text Track of 5th PVUW Challenge: 3rd Method","ref_index":21,"is_internal_anchor":false},{"citing_arxiv_id":"2604.18665","citing_title":"APRVOS: 1st Place Winner of 5th PVUW MeViS-Audio Track","ref_index":26,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/Y3SDCMFMWYSRI7NPUAA5X4LFT3","json":"https://pith.science/pith/Y3SDCMFMWYSRI7NPUAA5X4LFT3.json","graph_json":"https://pith.science/api/pith-number/Y3SDCMFMWYSRI7NPUAA5X4LFT3/graph.json","events_json":"https://pith.science/api/pith-number/Y3SDCMFMWYSRI7NPUAA5X4LFT3/events.json","paper":"https://pith.science/paper/Y3SDCMFM"},"agent_actions":{"view_html":"https://pith.science/pith/Y3SDCMFMWYSRI7NPUAA5X4LFT3","download_json":"https://pith.science/pith/Y3SDCMFMWYSRI7NPUAA5X4LFT3.json","view_paper":"https://pith.science/paper/Y3SDCMFM","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2404.07973&json=true","fetch_graph":"https://pith.science/api/pith-number/Y3SDCMFMWYSRI7NPUAA5X4LFT3/graph.json","fetch_events":"https://pith.science/api/pith-number/Y3SDCMFMWYSRI7NPUAA5X4LFT3/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/Y3SDCMFMWYSRI7NPUAA5X4LFT3/action/timestamp_anchor","attest_storage":"https://pith.science/pith/Y3SDCMFMWYSRI7NPUAA5X4LFT3/action/storage_attestation","attest_author":"https://pith.science/pith/Y3SDCMFMWYSRI7NPUAA5X4LFT3/action/author_attestation","sign_citation":"https://pith.science/pith/Y3SDCMFMWYSRI7NPUAA5X4LFT3/action/citation_signature","submit_replication":"https://pith.science/pith/Y3SDCMFMWYSRI7NPUAA5X4LFT3/action/replication_record"}},"created_at":"2026-07-05T08:07:00.068119+00:00","updated_at":"2026-07-05T08:07:00.068119+00:00"}