{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:YEM2XMUIZOKSUVPLSLYSCHXUL3","short_pith_number":"pith:YEM2XMUI","schema_version":"1.0","canonical_sha256":"c119abb288cb952a55eb92f1211ef45ed95ff769fbf3c3445976a819b7044bea","source":{"kind":"arxiv","id":"2509.07969","version":1},"attestation_state":"computed","paper":{"title":"Mini-o3: Scaling Up Reasoning Patterns and Interaction Turns for Visual Search","license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","headline":"","cross_cats":["cs.AI","cs.CL"],"primary_cat":"cs.CV","authors_text":"Hengshuang Zhao, Junyi Li, Tao Liu, Tianjian Li, Wei Li, Xin Lai","submitted_at":"2025-09-09T17:54:21Z","abstract_excerpt":"Recent advances in large multimodal models have leveraged image-based tools with reinforcement learning to tackle visual problems. However, existing open-source approaches often exhibit monotonous reasoning patterns and allow only a limited number of interaction turns, making them inadequate for difficult tasks that require trial-and-error exploration. In this work, we address this limitation by scaling up tool-based interactions and introduce Mini-o3, a system that executes deep, multi-turn reasoning -- spanning tens of steps -- and achieves state-of-the-art performance on challenging visual "},"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":"2509.07969","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","primary_cat":"cs.CV","submitted_at":"2025-09-09T17:54:21Z","cross_cats_sorted":["cs.AI","cs.CL"],"title_canon_sha256":"3dba4a2a14cbde4ca3b92e038522d8b92c7f9164b0fceff9668ca1ffc2c44439","abstract_canon_sha256":"49fc778abed99a546d7d53a16451449d3cb38981b22a4f4a3f92779e1d8afa9e"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-18T01:12:17.797291Z","signature_b64":"kQN4K5wjxAJk7iGkNzRGzxmSvKsg0sS7TMH3p1RvSO5jmzN4upcpcmhWLZiAZwEUh0KpSQAC/75Wk+xnb+LDBA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"c119abb288cb952a55eb92f1211ef45ed95ff769fbf3c3445976a819b7044bea","last_reissued_at":"2026-05-18T01:12:17.796873Z","signature_status":"signed_v1","first_computed_at":"2026-05-18T01:12:17.796873Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Mini-o3: Scaling Up Reasoning Patterns and Interaction Turns for Visual Search","license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","headline":"","cross_cats":["cs.AI","cs.CL"],"primary_cat":"cs.CV","authors_text":"Hengshuang Zhao, Junyi Li, Tao Liu, Tianjian Li, Wei Li, Xin Lai","submitted_at":"2025-09-09T17:54:21Z","abstract_excerpt":"Recent advances in large multimodal models have leveraged image-based tools with reinforcement learning to tackle visual problems. However, existing open-source approaches often exhibit monotonous reasoning patterns and allow only a limited number of interaction turns, making them inadequate for difficult tasks that require trial-and-error exploration. In this work, we address this limitation by scaling up tool-based interactions and introduce Mini-o3, a system that executes deep, multi-turn reasoning -- spanning tens of steps -- and achieves state-of-the-art performance on challenging visual "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2509.07969","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":""},"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":"2509.07969","created_at":"2026-05-18T01:12:17.796926+00:00"},{"alias_kind":"arxiv_version","alias_value":"2509.07969v1","created_at":"2026-05-18T01:12:17.796926+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2509.07969","created_at":"2026-05-18T01:12:17.796926+00:00"},{"alias_kind":"pith_short_12","alias_value":"YEM2XMUIZOKS","created_at":"2026-05-18T12:33:37.589309+00:00"},{"alias_kind":"pith_short_16","alias_value":"YEM2XMUIZOKSUVPL","created_at":"2026-05-18T12:33:37.589309+00:00"},{"alias_kind":"pith_short_8","alias_value":"YEM2XMUI","created_at":"2026-05-18T12:33:37.589309+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":18,"internal_anchor_count":18,"sample":[{"citing_arxiv_id":"2511.19820","citing_title":"CropVLM: Learning to Zoom for Fine-Grained Vision-Language Perception","ref_index":9,"is_internal_anchor":true},{"citing_arxiv_id":"2511.19972","citing_title":"Boosting Reasoning in Large Multimodal Models via Activation Replay","ref_index":18,"is_internal_anchor":true},{"citing_arxiv_id":"2512.16918","citing_title":"AdaTooler-V: Adaptive Tool-Use for Images and Videos","ref_index":25,"is_internal_anchor":true},{"citing_arxiv_id":"2601.15724","citing_title":"VideoThinker: Building Agentic VideoLLMs with LLM-Guided Tool Reasoning","ref_index":12,"is_internal_anchor":true},{"citing_arxiv_id":"2511.05271","citing_title":"DeepEyesV2: Toward Agentic Multimodal Model","ref_index":23,"is_internal_anchor":true},{"citing_arxiv_id":"2603.27494","citing_title":"Learning to Focus and Precise Cropping: A Reinforcement Learning Framework with Information Gaps and Grounding Loss for MLLMs","ref_index":14,"is_internal_anchor":true},{"citing_arxiv_id":"2604.02794","citing_title":"CharTool: Tool-Integrated Visual Reasoning for Chart Understanding","ref_index":22,"is_internal_anchor":true},{"citing_arxiv_id":"2604.03074","citing_title":"Speaker-Reasoner: Scaling Interaction Turns and Reasoning Patterns for Timestamped Speaker-Attributed ASR","ref_index":25,"is_internal_anchor":true},{"citing_arxiv_id":"2604.14125","citing_title":"HiVLA: A Visual-Grounded-Centric Hierarchical Embodied Manipulation System","ref_index":21,"is_internal_anchor":true},{"citing_arxiv_id":"2605.09146","citing_title":"Beyond Thinking: Imagining in 360$^\\circ$ for Humanoid Visual Search","ref_index":39,"is_internal_anchor":true},{"citing_arxiv_id":"2605.04304","citing_title":"Hierarchical Visual Agent: Managing Contexts in Joint Image-Text Space for Advanced Chart Reasoning","ref_index":10,"is_internal_anchor":true},{"citing_arxiv_id":"2604.19945","citing_title":"Visual Reasoning through Tool-supervised Reinforcement Learning","ref_index":9,"is_internal_anchor":true},{"citing_arxiv_id":"2604.06777","citing_title":"Walk the Talk: Bridging the Reasoning-Action Gap for Thinking with Images via Multimodal Agentic Policy Optimization","ref_index":7,"is_internal_anchor":true},{"citing_arxiv_id":"2604.08545","citing_title":"Act Wisely: Cultivating Meta-Cognitive Tool Use in Agentic Multimodal Models","ref_index":12,"is_internal_anchor":true},{"citing_arxiv_id":"2604.06912","citing_title":"Q-Zoom: Query-Aware Adaptive Perception for Efficient Multimodal Large Language Models","ref_index":6,"is_internal_anchor":true},{"citing_arxiv_id":"2604.14125","citing_title":"HiVLA: A Visual-Grounded-Centric Hierarchical Embodied Manipulation System","ref_index":21,"is_internal_anchor":true},{"citing_arxiv_id":"2604.16858","citing_title":"Q-DeepSight: Incentivizing Thinking with Images for Image Quality Assessment and Refinement","ref_index":17,"is_internal_anchor":true},{"citing_arxiv_id":"2605.04304","citing_title":"Hierarchical Visual Agent: Managing Contexts in Joint Image-Text Space for Advanced Chart Reasoning","ref_index":37,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/YEM2XMUIZOKSUVPLSLYSCHXUL3","json":"https://pith.science/pith/YEM2XMUIZOKSUVPLSLYSCHXUL3.json","graph_json":"https://pith.science/api/pith-number/YEM2XMUIZOKSUVPLSLYSCHXUL3/graph.json","events_json":"https://pith.science/api/pith-number/YEM2XMUIZOKSUVPLSLYSCHXUL3/events.json","paper":"https://pith.science/paper/YEM2XMUI"},"agent_actions":{"view_html":"https://pith.science/pith/YEM2XMUIZOKSUVPLSLYSCHXUL3","download_json":"https://pith.science/pith/YEM2XMUIZOKSUVPLSLYSCHXUL3.json","view_paper":"https://pith.science/paper/YEM2XMUI","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2509.07969&json=true","fetch_graph":"https://pith.science/api/pith-number/YEM2XMUIZOKSUVPLSLYSCHXUL3/graph.json","fetch_events":"https://pith.science/api/pith-number/YEM2XMUIZOKSUVPLSLYSCHXUL3/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/YEM2XMUIZOKSUVPLSLYSCHXUL3/action/timestamp_anchor","attest_storage":"https://pith.science/pith/YEM2XMUIZOKSUVPLSLYSCHXUL3/action/storage_attestation","attest_author":"https://pith.science/pith/YEM2XMUIZOKSUVPLSLYSCHXUL3/action/author_attestation","sign_citation":"https://pith.science/pith/YEM2XMUIZOKSUVPLSLYSCHXUL3/action/citation_signature","submit_replication":"https://pith.science/pith/YEM2XMUIZOKSUVPLSLYSCHXUL3/action/replication_record"}},"created_at":"2026-05-18T01:12:17.796926+00:00","updated_at":"2026-05-18T01:12:17.796926+00:00"}