{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:QVM56BKUITLL6KHPMWP2GAWANZ","short_pith_number":"pith:QVM56BKU","schema_version":"1.0","canonical_sha256":"8559df055444d6bf28ef659fa302c06e42f1f0171c16f931c3789cc379bc3dfe","source":{"kind":"arxiv","id":"2507.01016","version":1},"attestation_state":"computed","paper":{"title":"VQ-VLA: Improving Vision-Language-Action Models via Scaling Vector-Quantized Action Tokenizers","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":["cs.CV"],"primary_cat":"cs.RO","authors_text":"Hao-Shu Fang, Haoyi Zhu, Jiange Yang, Mingyu Liu, Tong He, Yating Wang","submitted_at":"2025-07-01T17:59:44Z","abstract_excerpt":"In this paper, we introduce an innovative vector quantization based action tokenizer built upon the largest-scale action trajectory dataset to date, leveraging over 100 times more data than previous approaches. This extensive dataset enables our tokenizer to capture rich spatiotemporal dynamics, resulting in a model that not only accelerates inference but also generates smoother and more coherent action outputs. Once trained, the tokenizer can be seamlessly adapted to a wide range of downstream tasks in a zero-shot manner, from short-horizon reactive behaviors to long-horizon planning. A key f"},"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":"2507.01016","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.RO","submitted_at":"2025-07-01T17:59:44Z","cross_cats_sorted":["cs.CV"],"title_canon_sha256":"b49b2f5162e514afd61e97db36cb78f2d10797c86f8a0b7b14144a79b7ca5a05","abstract_canon_sha256":"9b2368b18b767b43f6f85b76fffe092b5ce2d4e6a02b61bdf9acd3c172541768"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:30:19.618993Z","signature_b64":"X4QClZHPA+q2myyv1TLJivKxVpy/xvIJY6gCGW4MruLYHDcpANsSEPBkEm//vVHk7w8IyqSBVSaLkI6KvqQKCw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"8559df055444d6bf28ef659fa302c06e42f1f0171c16f931c3789cc379bc3dfe","last_reissued_at":"2026-07-05T11:30:19.618547Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:30:19.618547Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"VQ-VLA: Improving Vision-Language-Action Models via Scaling Vector-Quantized Action Tokenizers","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":["cs.CV"],"primary_cat":"cs.RO","authors_text":"Hao-Shu Fang, Haoyi Zhu, Jiange Yang, Mingyu Liu, Tong He, Yating Wang","submitted_at":"2025-07-01T17:59:44Z","abstract_excerpt":"In this paper, we introduce an innovative vector quantization based action tokenizer built upon the largest-scale action trajectory dataset to date, leveraging over 100 times more data than previous approaches. This extensive dataset enables our tokenizer to capture rich spatiotemporal dynamics, resulting in a model that not only accelerates inference but also generates smoother and more coherent action outputs. Once trained, the tokenizer can be seamlessly adapted to a wide range of downstream tasks in a zero-shot manner, from short-horizon reactive behaviors to long-horizon planning. A key f"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2507.01016","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/2507.01016/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":"2507.01016","created_at":"2026-07-05T11:30:19.618602+00:00"},{"alias_kind":"arxiv_version","alias_value":"2507.01016v1","created_at":"2026-07-05T11:30:19.618602+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2507.01016","created_at":"2026-07-05T11:30:19.618602+00:00"},{"alias_kind":"pith_short_12","alias_value":"QVM56BKUITLL","created_at":"2026-07-05T11:30:19.618602+00:00"},{"alias_kind":"pith_short_16","alias_value":"QVM56BKUITLL6KHP","created_at":"2026-07-05T11:30:19.618602+00:00"},{"alias_kind":"pith_short_8","alias_value":"QVM56BKU","created_at":"2026-07-05T11:30:19.618602+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":9,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.22480","citing_title":"ARP: Enhancing Quantized Skill Abstractions via Visual Alignment and Iterative Refinement for Robotic Manipulation","ref_index":34,"is_internal_anchor":false},{"citing_arxiv_id":"2606.21372","citing_title":"NAC: Neural Action Codec for Vision-Language-Action Models","ref_index":25,"is_internal_anchor":false},{"citing_arxiv_id":"2605.12369","citing_title":"GuidedVLA: Specifying Task-Relevant Factors via Plug-and-Play Action Attention Specialization","ref_index":85,"is_internal_anchor":false},{"citing_arxiv_id":"2605.22671","citing_title":"From Abstraction to Instantiation: Learning Behavioral Representation for Vision-Language-Action Model","ref_index":25,"is_internal_anchor":false},{"citing_arxiv_id":"2605.22671","citing_title":"From Abstraction to Instantiation: Learning Behavioral Representation for Vision-Language-Action Model","ref_index":25,"is_internal_anchor":false},{"citing_arxiv_id":"2602.19710","citing_title":"PoseVLA: Universal Pose Pretraining for Generalizable Vision-Language-Action Policies","ref_index":41,"is_internal_anchor":false},{"citing_arxiv_id":"2602.12978","citing_title":"Learning Native Continuation for Action Chunking Flow Policies","ref_index":35,"is_internal_anchor":false},{"citing_arxiv_id":"2511.18085","citing_title":"Continually Evolving Skill Knowledge in Vision Language Action Model","ref_index":41,"is_internal_anchor":false},{"citing_arxiv_id":"2605.12369","citing_title":"GuidedVLA: Specifying Task-Relevant Factors via Plug-and-Play Action Attention Specialization","ref_index":85,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/QVM56BKUITLL6KHPMWP2GAWANZ","json":"https://pith.science/pith/QVM56BKUITLL6KHPMWP2GAWANZ.json","graph_json":"https://pith.science/api/pith-number/QVM56BKUITLL6KHPMWP2GAWANZ/graph.json","events_json":"https://pith.science/api/pith-number/QVM56BKUITLL6KHPMWP2GAWANZ/events.json","paper":"https://pith.science/paper/QVM56BKU"},"agent_actions":{"view_html":"https://pith.science/pith/QVM56BKUITLL6KHPMWP2GAWANZ","download_json":"https://pith.science/pith/QVM56BKUITLL6KHPMWP2GAWANZ.json","view_paper":"https://pith.science/paper/QVM56BKU","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2507.01016&json=true","fetch_graph":"https://pith.science/api/pith-number/QVM56BKUITLL6KHPMWP2GAWANZ/graph.json","fetch_events":"https://pith.science/api/pith-number/QVM56BKUITLL6KHPMWP2GAWANZ/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/QVM56BKUITLL6KHPMWP2GAWANZ/action/timestamp_anchor","attest_storage":"https://pith.science/pith/QVM56BKUITLL6KHPMWP2GAWANZ/action/storage_attestation","attest_author":"https://pith.science/pith/QVM56BKUITLL6KHPMWP2GAWANZ/action/author_attestation","sign_citation":"https://pith.science/pith/QVM56BKUITLL6KHPMWP2GAWANZ/action/citation_signature","submit_replication":"https://pith.science/pith/QVM56BKUITLL6KHPMWP2GAWANZ/action/replication_record"}},"created_at":"2026-07-05T11:30:19.618602+00:00","updated_at":"2026-07-05T11:30:19.618602+00:00"}