{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:HV4DE7724GXOBHSCOGD4DSLCJY","short_pith_number":"pith:HV4DE772","schema_version":"1.0","canonical_sha256":"3d78327ffae1aee09e427187c1c9624e0938aedb42163f57facc5f613077e830","source":{"kind":"arxiv","id":"2505.20289","version":2},"attestation_state":"computed","paper":{"title":"VisualToolAgent (VisTA): A Reinforcement Learning Framework for Visual Tool Selection","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Anirudh Sundara Rajan, Haohan Wang, Junjie Hu, Wen Xiao, Yong Jae Lee, Yuyang Ji, Zefan Cai, Zeyi Huang","submitted_at":"2025-05-26T17:59:17Z","abstract_excerpt":"We introduce VisTA, a new reinforcement learning framework that empowers visual agents to dynamically explore, select, and combine tools from a diverse library based on empirical performance. Existing methods for tool-augmented reasoning either rely on training-free prompting or large-scale fine-tuning; both lack active tool exploration and typically assume limited tool diversity, and fine-tuning methods additionally demand extensive human supervision. In contrast, VisTA leverages end-to-end reinforcement learning to iteratively refine sophisticated, query-specific tool selection strategies, u"},"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":"2505.20289","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2025-05-26T17:59:17Z","cross_cats_sorted":[],"title_canon_sha256":"6d4fa7d73198168034b8420872212384d1dbd6c59a15dcede1989b4626a33ee0","abstract_canon_sha256":"75c87d0353aa617513a8e230238e6ff17f833ca768c758642b6ec35391fa8cce"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:39:54.368398Z","signature_b64":"6xLQU2ZtojAZa2UB7HqaVPCHSgxclUViCwn0CiT6zq9KhxHN5uFLGrEMM7Kph+ly/RvESGtkUHFzASY4QjmPBg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"3d78327ffae1aee09e427187c1c9624e0938aedb42163f57facc5f613077e830","last_reissued_at":"2026-07-05T11:39:54.367857Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:39:54.367857Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"VisualToolAgent (VisTA): A Reinforcement Learning Framework for Visual Tool Selection","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Anirudh Sundara Rajan, Haohan Wang, Junjie Hu, Wen Xiao, Yong Jae Lee, Yuyang Ji, Zefan Cai, Zeyi Huang","submitted_at":"2025-05-26T17:59:17Z","abstract_excerpt":"We introduce VisTA, a new reinforcement learning framework that empowers visual agents to dynamically explore, select, and combine tools from a diverse library based on empirical performance. Existing methods for tool-augmented reasoning either rely on training-free prompting or large-scale fine-tuning; both lack active tool exploration and typically assume limited tool diversity, and fine-tuning methods additionally demand extensive human supervision. In contrast, VisTA leverages end-to-end reinforcement learning to iteratively refine sophisticated, query-specific tool selection strategies, u"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2505.20289","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/2505.20289/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":"2505.20289","created_at":"2026-07-05T11:39:54.367919+00:00"},{"alias_kind":"arxiv_version","alias_value":"2505.20289v2","created_at":"2026-07-05T11:39:54.367919+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2505.20289","created_at":"2026-07-05T11:39:54.367919+00:00"},{"alias_kind":"pith_short_12","alias_value":"HV4DE7724GXO","created_at":"2026-07-05T11:39:54.367919+00:00"},{"alias_kind":"pith_short_16","alias_value":"HV4DE7724GXOBHSC","created_at":"2026-07-05T11:39:54.367919+00:00"},{"alias_kind":"pith_short_8","alias_value":"HV4DE772","created_at":"2026-07-05T11:39:54.367919+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":7,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.26196","citing_title":"From Structure to Synergy: A Survey of Vision-Language Perception Paradigm Evolution in Multimodal Large Language Models","ref_index":214,"is_internal_anchor":false},{"citing_arxiv_id":"2605.26861","citing_title":"REVERSE: Reinforcing Evidence Verification and Search for Agentic Image geo-localization","ref_index":30,"is_internal_anchor":false},{"citing_arxiv_id":"2606.00562","citing_title":"DeepLatent: Think with Images via Parallel Latent Visual Reasoning","ref_index":21,"is_internal_anchor":false},{"citing_arxiv_id":"2509.02547","citing_title":"The Landscape of Agentic Reinforcement Learning for LLMs: A Survey","ref_index":252,"is_internal_anchor":false},{"citing_arxiv_id":"2509.24251","citing_title":"Latent Visual Reasoning","ref_index":12,"is_internal_anchor":false},{"citing_arxiv_id":"2605.12163","citing_title":"Self-Consistent Latent Reasoning: Long Latent Sequence Reasoning for Vision-Language Model","ref_index":18,"is_internal_anchor":false},{"citing_arxiv_id":"2605.12163","citing_title":"Self-Consistent Latent Reasoning: Long Latent Sequence Reasoning for Vision-Language Model","ref_index":18,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/HV4DE7724GXOBHSCOGD4DSLCJY","json":"https://pith.science/pith/HV4DE7724GXOBHSCOGD4DSLCJY.json","graph_json":"https://pith.science/api/pith-number/HV4DE7724GXOBHSCOGD4DSLCJY/graph.json","events_json":"https://pith.science/api/pith-number/HV4DE7724GXOBHSCOGD4DSLCJY/events.json","paper":"https://pith.science/paper/HV4DE772"},"agent_actions":{"view_html":"https://pith.science/pith/HV4DE7724GXOBHSCOGD4DSLCJY","download_json":"https://pith.science/pith/HV4DE7724GXOBHSCOGD4DSLCJY.json","view_paper":"https://pith.science/paper/HV4DE772","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2505.20289&json=true","fetch_graph":"https://pith.science/api/pith-number/HV4DE7724GXOBHSCOGD4DSLCJY/graph.json","fetch_events":"https://pith.science/api/pith-number/HV4DE7724GXOBHSCOGD4DSLCJY/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/HV4DE7724GXOBHSCOGD4DSLCJY/action/timestamp_anchor","attest_storage":"https://pith.science/pith/HV4DE7724GXOBHSCOGD4DSLCJY/action/storage_attestation","attest_author":"https://pith.science/pith/HV4DE7724GXOBHSCOGD4DSLCJY/action/author_attestation","sign_citation":"https://pith.science/pith/HV4DE7724GXOBHSCOGD4DSLCJY/action/citation_signature","submit_replication":"https://pith.science/pith/HV4DE7724GXOBHSCOGD4DSLCJY/action/replication_record"}},"created_at":"2026-07-05T11:39:54.367919+00:00","updated_at":"2026-07-05T11:39:54.367919+00:00"}