{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:QSPMIAV2PFEOFX7C7L3N4MNKYL","short_pith_number":"pith:QSPMIAV2","schema_version":"1.0","canonical_sha256":"849ec402ba7948e2dfe2faf6de31aac2d29052ccf88dfe7d7778be90804c8f33","source":{"kind":"arxiv","id":"2507.13348","version":1},"attestation_state":"computed","paper":{"title":"VisionThink: Smart and Efficient Vision Language Model via Reinforcement Learning","license":"http://creativecommons.org/licenses/by-sa/4.0/","headline":"","cross_cats":["cs.AI","cs.CL","cs.LG"],"primary_cat":"cs.CV","authors_text":"Bei Yu, Hengshuang Zhao, Jiaya Jia, Junyi Li, Senqiao Yang, Xin Lai","submitted_at":"2025-07-17T17:59:55Z","abstract_excerpt":"Recent advancements in vision-language models (VLMs) have improved performance by increasing the number of visual tokens, which are often significantly longer than text tokens. However, we observe that most real-world scenarios do not require such an extensive number of visual tokens. While the performance drops significantly in a small subset of OCR-related tasks, models still perform accurately in most other general VQA tasks with only 1/4 resolution. Therefore, we propose to dynamically process distinct samples with different resolutions, and present a new paradigm for visual token compress"},"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":"2507.13348","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.CV","submitted_at":"2025-07-17T17:59:55Z","cross_cats_sorted":["cs.AI","cs.CL","cs.LG"],"title_canon_sha256":"1fa5841651be76533687881151ee3966135e8c7e6b59f05948f29bde46b273e8","abstract_canon_sha256":"9b3fd152feacb0ff29d96fe568746fe22a603e17ea85c6212866b569b8486734"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:38:59.605424Z","signature_b64":"YbsNvB+2izvXuLMXB7LZEeK4GViKIXXDC7O/iTILVHCo/BFA0PiBd2sjz92SrSidX5De3lWkYg8Go5UMJ7uzBQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"849ec402ba7948e2dfe2faf6de31aac2d29052ccf88dfe7d7778be90804c8f33","last_reissued_at":"2026-07-05T11:38:59.604960Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:38:59.604960Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"VisionThink: Smart and Efficient Vision Language Model via Reinforcement Learning","license":"http://creativecommons.org/licenses/by-sa/4.0/","headline":"","cross_cats":["cs.AI","cs.CL","cs.LG"],"primary_cat":"cs.CV","authors_text":"Bei Yu, Hengshuang Zhao, Jiaya Jia, Junyi Li, Senqiao Yang, Xin Lai","submitted_at":"2025-07-17T17:59:55Z","abstract_excerpt":"Recent advancements in vision-language models (VLMs) have improved performance by increasing the number of visual tokens, which are often significantly longer than text tokens. However, we observe that most real-world scenarios do not require such an extensive number of visual tokens. While the performance drops significantly in a small subset of OCR-related tasks, models still perform accurately in most other general VQA tasks with only 1/4 resolution. Therefore, we propose to dynamically process distinct samples with different resolutions, and present a new paradigm for visual token compress"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2507.13348","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/2507.13348/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":"2507.13348","created_at":"2026-07-05T11:38:59.605011+00:00"},{"alias_kind":"arxiv_version","alias_value":"2507.13348v1","created_at":"2026-07-05T11:38:59.605011+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2507.13348","created_at":"2026-07-05T11:38:59.605011+00:00"},{"alias_kind":"pith_short_12","alias_value":"QSPMIAV2PFEO","created_at":"2026-07-05T11:38:59.605011+00:00"},{"alias_kind":"pith_short_16","alias_value":"QSPMIAV2PFEOFX7C","created_at":"2026-07-05T11:38:59.605011+00:00"},{"alias_kind":"pith_short_8","alias_value":"QSPMIAV2","created_at":"2026-07-05T11:38:59.605011+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":8,"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":216,"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":59,"is_internal_anchor":false},{"citing_arxiv_id":"2606.31599","citing_title":"Token-Sparse Medical Multimodal Reasoning via Dual-Stream Reinforcement Learning","ref_index":31,"is_internal_anchor":false},{"citing_arxiv_id":"2509.07969","citing_title":"Mini-o3: Scaling Up Reasoning Patterns and Interaction Turns for Visual Search","ref_index":43,"is_internal_anchor":false},{"citing_arxiv_id":"2604.03179","citing_title":"Understanding the Role of Hallucination in Reinforcement Post-Training of Multimodal Reasoning Models","ref_index":36,"is_internal_anchor":false},{"citing_arxiv_id":"2605.06121","citing_title":"Pest-Thinker: Learning to Think and Reason like Entomologists via Reinforcement Learning","ref_index":37,"is_internal_anchor":false},{"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":65,"is_internal_anchor":false},{"citing_arxiv_id":"2604.04500","citing_title":"Saliency-R1: Enforcing Interpretable and Faithful Vision-language Reasoning via Saliency-map Alignment Reward","ref_index":80,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/QSPMIAV2PFEOFX7C7L3N4MNKYL","json":"https://pith.science/pith/QSPMIAV2PFEOFX7C7L3N4MNKYL.json","graph_json":"https://pith.science/api/pith-number/QSPMIAV2PFEOFX7C7L3N4MNKYL/graph.json","events_json":"https://pith.science/api/pith-number/QSPMIAV2PFEOFX7C7L3N4MNKYL/events.json","paper":"https://pith.science/paper/QSPMIAV2"},"agent_actions":{"view_html":"https://pith.science/pith/QSPMIAV2PFEOFX7C7L3N4MNKYL","download_json":"https://pith.science/pith/QSPMIAV2PFEOFX7C7L3N4MNKYL.json","view_paper":"https://pith.science/paper/QSPMIAV2","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2507.13348&json=true","fetch_graph":"https://pith.science/api/pith-number/QSPMIAV2PFEOFX7C7L3N4MNKYL/graph.json","fetch_events":"https://pith.science/api/pith-number/QSPMIAV2PFEOFX7C7L3N4MNKYL/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/QSPMIAV2PFEOFX7C7L3N4MNKYL/action/timestamp_anchor","attest_storage":"https://pith.science/pith/QSPMIAV2PFEOFX7C7L3N4MNKYL/action/storage_attestation","attest_author":"https://pith.science/pith/QSPMIAV2PFEOFX7C7L3N4MNKYL/action/author_attestation","sign_citation":"https://pith.science/pith/QSPMIAV2PFEOFX7C7L3N4MNKYL/action/citation_signature","submit_replication":"https://pith.science/pith/QSPMIAV2PFEOFX7C7L3N4MNKYL/action/replication_record"}},"created_at":"2026-07-05T11:38:59.605011+00:00","updated_at":"2026-07-05T11:38:59.605011+00:00"}