{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:Z3YQ75XKPRDJ7KZYYHJFYZLSNK","short_pith_number":"pith:Z3YQ75XK","schema_version":"1.0","canonical_sha256":"cef10ff6ea7c469fab38c1d25c65726a938ccd739b7b27afa0cf7a139dc3593e","source":{"kind":"arxiv","id":"2505.14246","version":1},"attestation_state":"computed","paper":{"title":"Visual Agentic Reinforcement Fine-Tuning","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CV","authors_text":"Dahua Lin, Haodong Duan, Jiaqi Wang, Xiaoyi Dong, Yuhang Cao, Yuhang Zang, Yushan Zou, Zijian Liang, Ziyu Liu","submitted_at":"2025-05-20T11:59:25Z","abstract_excerpt":"A key trend in Large Reasoning Models (e.g., OpenAI's o3) is the native agentic ability to use external tools such as web browsers for searching and writing/executing code for image manipulation to think with images. In the open-source research community, while significant progress has been made in language-only agentic abilities such as function calling and tool integration, the development of multi-modal agentic capabilities that involve truly thinking with images, and their corresponding benchmarks, are still less explored. This work highlights the effectiveness of Visual Agentic Reinforcem"},"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":"2505.14246","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2025-05-20T11:59:25Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"ace08cc8d5f61bcc13c6609beddab3c1f3438967cc5f6955511173bbb7f422b5","abstract_canon_sha256":"60cc85778631bf0d63feb621399bb6f5fc8f349486c8ecbdb156df20220fa22e"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:05:58.381949Z","signature_b64":"PrXsSmaH1cmQiWPLv46a+7xwGn02IWfj4UkmLjpiDMam6srVFCw7TbXVYDQUv62XZk4BA5HWp8bFYp9Q21U+AA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"cef10ff6ea7c469fab38c1d25c65726a938ccd739b7b27afa0cf7a139dc3593e","last_reissued_at":"2026-07-05T11:05:58.381465Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:05:58.381465Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Visual Agentic Reinforcement Fine-Tuning","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CV","authors_text":"Dahua Lin, Haodong Duan, Jiaqi Wang, Xiaoyi Dong, Yuhang Cao, Yuhang Zang, Yushan Zou, Zijian Liang, Ziyu Liu","submitted_at":"2025-05-20T11:59:25Z","abstract_excerpt":"A key trend in Large Reasoning Models (e.g., OpenAI's o3) is the native agentic ability to use external tools such as web browsers for searching and writing/executing code for image manipulation to think with images. In the open-source research community, while significant progress has been made in language-only agentic abilities such as function calling and tool integration, the development of multi-modal agentic capabilities that involve truly thinking with images, and their corresponding benchmarks, are still less explored. This work highlights the effectiveness of Visual Agentic Reinforcem"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2505.14246","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/2505.14246/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":"2505.14246","created_at":"2026-07-05T11:05:58.381518+00:00"},{"alias_kind":"arxiv_version","alias_value":"2505.14246v1","created_at":"2026-07-05T11:05:58.381518+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2505.14246","created_at":"2026-07-05T11:05:58.381518+00:00"},{"alias_kind":"pith_short_12","alias_value":"Z3YQ75XKPRDJ","created_at":"2026-07-05T11:05:58.381518+00:00"},{"alias_kind":"pith_short_16","alias_value":"Z3YQ75XKPRDJ7KZY","created_at":"2026-07-05T11:05:58.381518+00:00"},{"alias_kind":"pith_short_8","alias_value":"Z3YQ75XK","created_at":"2026-07-05T11:05:58.381518+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":17,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.15932","citing_title":"Beyond NL2Code: A Structured Survey of Multimodal Code Intelligence","ref_index":43,"is_internal_anchor":false},{"citing_arxiv_id":"2606.05784","citing_title":"TAPO: Tool-Aware Policy Optimization via Credit Transfer for Multimodal Search Agents","ref_index":8,"is_internal_anchor":false},{"citing_arxiv_id":"2606.31504","citing_title":"SimpleSearch-VL: A Simple Recipe for Multimodal Agentic Deep Search","ref_index":79,"is_internal_anchor":false},{"citing_arxiv_id":"2605.22177","citing_title":"Maestro: Reinforcement Learning to Orchestrate Hierarchical Model-Skill Ensembles","ref_index":27,"is_internal_anchor":false},{"citing_arxiv_id":"2605.09860","citing_title":"When to Re-Commit: Temporal Abstraction Discovery for Long-Horizon Vision-Language Reasoning","ref_index":34,"is_internal_anchor":false},{"citing_arxiv_id":"2605.09860","citing_title":"When to Re-Commit: Temporal Abstraction Discovery for Long-Horizon Vision-Language Reasoning","ref_index":34,"is_internal_anchor":false},{"citing_arxiv_id":"2605.15792","citing_title":"Reversing the Flow: Generation-to-Understanding Synergy in Large Multimodal Models","ref_index":23,"is_internal_anchor":false},{"citing_arxiv_id":"2512.16300","citing_title":"Code-in-the-Loop Forensics: Agentic Tool Use for Image Forgery Detection","ref_index":22,"is_internal_anchor":false},{"citing_arxiv_id":"2511.05271","citing_title":"DeepEyesV2: Toward Agentic Multimodal Model","ref_index":34,"is_internal_anchor":false},{"citing_arxiv_id":"2604.02794","citing_title":"CharTool: Tool-Integrated Visual Reasoning for Chart Understanding","ref_index":34,"is_internal_anchor":false},{"citing_arxiv_id":"2605.09860","citing_title":"When to Re-Commit: Temporal Abstraction Discovery for Long-Horizon Vision-Language Reasoning","ref_index":34,"is_internal_anchor":false},{"citing_arxiv_id":"2604.09508","citing_title":"VISOR: Agentic Visual Retrieval-Augmented Generation via Iterative Search and Over-horizon Reasoning","ref_index":19,"is_internal_anchor":false},{"citing_arxiv_id":"2604.14029","citing_title":"POINTS-Seeker: An Open Recipe for Multimodal Search Agents with Visual Memory Management","ref_index":29,"is_internal_anchor":false},{"citing_arxiv_id":"2604.14520","citing_title":"Chain of Modality: From Static Fusion to Dynamic Orchestration in Omni-MLLMs","ref_index":23,"is_internal_anchor":false},{"citing_arxiv_id":"2604.19857","citing_title":"Rethinking Reinforcement Fine-Tuning in LVLM: Convergence, Reward Decomposition, and Generalization","ref_index":1,"is_internal_anchor":false},{"citing_arxiv_id":"2604.21409","citing_title":"S1-VL: Scientific Multimodal Reasoning Model with Thinking-with-Images","ref_index":10,"is_internal_anchor":false},{"citing_arxiv_id":"2605.02730","citing_title":"Perceptual Flow Network for Visually Grounded Reasoning","ref_index":32,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/Z3YQ75XKPRDJ7KZYYHJFYZLSNK","json":"https://pith.science/pith/Z3YQ75XKPRDJ7KZYYHJFYZLSNK.json","graph_json":"https://pith.science/api/pith-number/Z3YQ75XKPRDJ7KZYYHJFYZLSNK/graph.json","events_json":"https://pith.science/api/pith-number/Z3YQ75XKPRDJ7KZYYHJFYZLSNK/events.json","paper":"https://pith.science/paper/Z3YQ75XK"},"agent_actions":{"view_html":"https://pith.science/pith/Z3YQ75XKPRDJ7KZYYHJFYZLSNK","download_json":"https://pith.science/pith/Z3YQ75XKPRDJ7KZYYHJFYZLSNK.json","view_paper":"https://pith.science/paper/Z3YQ75XK","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2505.14246&json=true","fetch_graph":"https://pith.science/api/pith-number/Z3YQ75XKPRDJ7KZYYHJFYZLSNK/graph.json","fetch_events":"https://pith.science/api/pith-number/Z3YQ75XKPRDJ7KZYYHJFYZLSNK/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/Z3YQ75XKPRDJ7KZYYHJFYZLSNK/action/timestamp_anchor","attest_storage":"https://pith.science/pith/Z3YQ75XKPRDJ7KZYYHJFYZLSNK/action/storage_attestation","attest_author":"https://pith.science/pith/Z3YQ75XKPRDJ7KZYYHJFYZLSNK/action/author_attestation","sign_citation":"https://pith.science/pith/Z3YQ75XKPRDJ7KZYYHJFYZLSNK/action/citation_signature","submit_replication":"https://pith.science/pith/Z3YQ75XKPRDJ7KZYYHJFYZLSNK/action/replication_record"}},"created_at":"2026-07-05T11:05:58.381518+00:00","updated_at":"2026-07-05T11:05:58.381518+00:00"}