{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:XKUS2V6K7DUZV3QYDDTWN7NF55","short_pith_number":"pith:XKUS2V6K","schema_version":"1.0","canonical_sha256":"baa92d57caf8e99aee1818e766fda5ef5b09c81a10cb00f32dfc5959ca3e1587","source":{"kind":"arxiv","id":"2505.13426","version":1},"attestation_state":"computed","paper":{"title":"G1: Bootstrapping Perception and Reasoning Abilities of Vision-Language Model via Reinforcement Learning","license":"http://creativecommons.org/licenses/by-sa/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Baobao Chang, Flood Sung, Hongcheng Gao, Liang Chen, Tianyu Liu, Xinyu Zhou, Yuxin Wu, Zhiqi Huang","submitted_at":"2025-05-19T17:54:39Z","abstract_excerpt":"Vision-Language Models (VLMs) excel in many direct multimodal tasks but struggle to translate this prowess into effective decision-making within interactive, visually rich environments like games. This ``knowing-doing'' gap significantly limits their potential as autonomous agents, as leading VLMs often performing badly in simple games. To address this, we introduce VLM-Gym, a curated reinforcement learning (RL) environment featuring diverse visual games with unified interfaces and adjustable, compositional difficulty, specifically designed for scalable multi-game parallel training. Leveraging"},"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":"2505.13426","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.CV","submitted_at":"2025-05-19T17:54:39Z","cross_cats_sorted":[],"title_canon_sha256":"5fcf76951319546f70ae635776896969c90e784e2b153856d6d06e1234129005","abstract_canon_sha256":"1ee1f2eb30885cb1ddded5e67f65327cc473fe18276c722ec9a18e3008a080ec"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:05:30.021080Z","signature_b64":"Am9ZNXw7D9PXjVnlVxoSFDAuj8wrQR8Zr0nKU3O/fozP6kuC0TWKsLH3N4tFlJvwi7ImX88EsmWBo83y1dTkCg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"baa92d57caf8e99aee1818e766fda5ef5b09c81a10cb00f32dfc5959ca3e1587","last_reissued_at":"2026-07-05T11:05:30.020608Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:05:30.020608Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"G1: Bootstrapping Perception and Reasoning Abilities of Vision-Language Model via Reinforcement Learning","license":"http://creativecommons.org/licenses/by-sa/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Baobao Chang, Flood Sung, Hongcheng Gao, Liang Chen, Tianyu Liu, Xinyu Zhou, Yuxin Wu, Zhiqi Huang","submitted_at":"2025-05-19T17:54:39Z","abstract_excerpt":"Vision-Language Models (VLMs) excel in many direct multimodal tasks but struggle to translate this prowess into effective decision-making within interactive, visually rich environments like games. This ``knowing-doing'' gap significantly limits their potential as autonomous agents, as leading VLMs often performing badly in simple games. To address this, we introduce VLM-Gym, a curated reinforcement learning (RL) environment featuring diverse visual games with unified interfaces and adjustable, compositional difficulty, specifically designed for scalable multi-game parallel training. Leveraging"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2505.13426","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/2505.13426/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":"2505.13426","created_at":"2026-07-05T11:05:30.020665+00:00"},{"alias_kind":"arxiv_version","alias_value":"2505.13426v1","created_at":"2026-07-05T11:05:30.020665+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2505.13426","created_at":"2026-07-05T11:05:30.020665+00:00"},{"alias_kind":"pith_short_12","alias_value":"XKUS2V6K7DUZ","created_at":"2026-07-05T11:05:30.020665+00:00"},{"alias_kind":"pith_short_16","alias_value":"XKUS2V6K7DUZV3QY","created_at":"2026-07-05T11:05:30.020665+00:00"},{"alias_kind":"pith_short_8","alias_value":"XKUS2V6K","created_at":"2026-07-05T11:05:30.020665+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":9,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.30217","citing_title":"Before Thinking, Learn to Decide: Proactive Routing for Efficient Visual Reasoning","ref_index":5,"is_internal_anchor":false},{"citing_arxiv_id":"2606.28266","citing_title":"RSICCLLM: A Multimodal Large Language Model for Remote Sensing Image Change Captioning","ref_index":7,"is_internal_anchor":false},{"citing_arxiv_id":"2508.11196","citing_title":"UAV-VL-R1: Generalizing Vision-Language Models via Supervised Fine-Tuning and Multi-Stage GRPO for UAV Visual Reasoning","ref_index":36,"is_internal_anchor":false},{"citing_arxiv_id":"2509.02547","citing_title":"The Landscape of Agentic Reinforcement Learning for LLMs: A Survey","ref_index":239,"is_internal_anchor":false},{"citing_arxiv_id":"2509.08827","citing_title":"A Survey of Reinforcement Learning for Large Reasoning Models","ref_index":61,"is_internal_anchor":false},{"citing_arxiv_id":"2603.15432","citing_title":"Gym-V: A Unified Vision Environment System for Agentic Vision Research","ref_index":3,"is_internal_anchor":false},{"citing_arxiv_id":"2605.09965","citing_title":"Towards Generalist Game Players: An Investigation of Foundation Models in the Game Multiverse","ref_index":25,"is_internal_anchor":false},{"citing_arxiv_id":"2605.09965","citing_title":"Towards Generalist Game Players: An Investigation of Foundation Models in the Game Multiverse","ref_index":25,"is_internal_anchor":false},{"citing_arxiv_id":"2605.00347","citing_title":"Odysseus: Scaling VLMs to 100+ Turn Decision-Making in Games via Reinforcement Learning","ref_index":74,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/XKUS2V6K7DUZV3QYDDTWN7NF55","json":"https://pith.science/pith/XKUS2V6K7DUZV3QYDDTWN7NF55.json","graph_json":"https://pith.science/api/pith-number/XKUS2V6K7DUZV3QYDDTWN7NF55/graph.json","events_json":"https://pith.science/api/pith-number/XKUS2V6K7DUZV3QYDDTWN7NF55/events.json","paper":"https://pith.science/paper/XKUS2V6K"},"agent_actions":{"view_html":"https://pith.science/pith/XKUS2V6K7DUZV3QYDDTWN7NF55","download_json":"https://pith.science/pith/XKUS2V6K7DUZV3QYDDTWN7NF55.json","view_paper":"https://pith.science/paper/XKUS2V6K","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2505.13426&json=true","fetch_graph":"https://pith.science/api/pith-number/XKUS2V6K7DUZV3QYDDTWN7NF55/graph.json","fetch_events":"https://pith.science/api/pith-number/XKUS2V6K7DUZV3QYDDTWN7NF55/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/XKUS2V6K7DUZV3QYDDTWN7NF55/action/timestamp_anchor","attest_storage":"https://pith.science/pith/XKUS2V6K7DUZV3QYDDTWN7NF55/action/storage_attestation","attest_author":"https://pith.science/pith/XKUS2V6K7DUZV3QYDDTWN7NF55/action/author_attestation","sign_citation":"https://pith.science/pith/XKUS2V6K7DUZV3QYDDTWN7NF55/action/citation_signature","submit_replication":"https://pith.science/pith/XKUS2V6K7DUZV3QYDDTWN7NF55/action/replication_record"}},"created_at":"2026-07-05T11:05:30.020665+00:00","updated_at":"2026-07-05T11:05:30.020665+00:00"}