{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:L6TZCUV5OYB5IU2FYWSKULNMAP","short_pith_number":"pith:L6TZCUV5","schema_version":"1.0","canonical_sha256":"5fa79152bd7603d45345c5a4aa2dac03e9554f5e0dc8878e4a9de747f10d5399","source":{"kind":"arxiv","id":"2606.29350","version":1},"attestation_state":"computed","paper":{"title":"Fast Enough to Act: Spatio-Temporal Visual Token Merging for Low-Latency Robotic VLMs and VLAs","license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CV","authors_text":"Gang Zhou, Jindong Wang, Junzhou Chen","submitted_at":"2026-06-28T11:42:55Z","abstract_excerpt":"Vision-language models and vision-language action models endow the robot with unprecedented capabilities. However, the input of video and high-resolution images yields a massive number of visual tokens, leading to extremely high inference latency and severely hindering the robot's real-time control. To break through this computational bottleneck, we propose ST-Merge, a plug-and-play, training-free framework that efficiently fuses redundant tokens directly during the visual encoding phase. By explicitly constructing 3D spatiotemporal coordinates, it employs a multi-queue parallel matching and w"},"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":"2606.29350","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","primary_cat":"cs.CV","submitted_at":"2026-06-28T11:42:55Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"9ec0558fda28c89f92226d6623f8176a296421042660202c89698ab372648b11","abstract_canon_sha256":"3ebd3f97994f542faedd87b916c47a52d9ad9ec014840e5c72b83bfe2b9e88e3"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-30T01:18:02.386241Z","signature_b64":"GqHeNAI+/f7dER0oRSsR0tyzejgD7iPwaoHCybYPSwUqT1aC3u43v+D5EGcxKQ6iwApq2zZ8fY65wH71bze9BQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"5fa79152bd7603d45345c5a4aa2dac03e9554f5e0dc8878e4a9de747f10d5399","last_reissued_at":"2026-06-30T01:18:02.385752Z","signature_status":"signed_v1","first_computed_at":"2026-06-30T01:18:02.385752Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Fast Enough to Act: Spatio-Temporal Visual Token Merging for Low-Latency Robotic VLMs and VLAs","license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CV","authors_text":"Gang Zhou, Jindong Wang, Junzhou Chen","submitted_at":"2026-06-28T11:42:55Z","abstract_excerpt":"Vision-language models and vision-language action models endow the robot with unprecedented capabilities. However, the input of video and high-resolution images yields a massive number of visual tokens, leading to extremely high inference latency and severely hindering the robot's real-time control. To break through this computational bottleneck, we propose ST-Merge, a plug-and-play, training-free framework that efficiently fuses redundant tokens directly during the visual encoding phase. By explicitly constructing 3D spatiotemporal coordinates, it employs a multi-queue parallel matching and w"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2606.29350","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/2606.29350/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":"2606.29350","created_at":"2026-06-30T01:18:02.385823+00:00"},{"alias_kind":"arxiv_version","alias_value":"2606.29350v1","created_at":"2026-06-30T01:18:02.385823+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2606.29350","created_at":"2026-06-30T01:18:02.385823+00:00"},{"alias_kind":"pith_short_12","alias_value":"L6TZCUV5OYB5","created_at":"2026-06-30T01:18:02.385823+00:00"},{"alias_kind":"pith_short_16","alias_value":"L6TZCUV5OYB5IU2F","created_at":"2026-06-30T01:18:02.385823+00:00"},{"alias_kind":"pith_short_8","alias_value":"L6TZCUV5","created_at":"2026-06-30T01:18:02.385823+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/L6TZCUV5OYB5IU2FYWSKULNMAP","json":"https://pith.science/pith/L6TZCUV5OYB5IU2FYWSKULNMAP.json","graph_json":"https://pith.science/api/pith-number/L6TZCUV5OYB5IU2FYWSKULNMAP/graph.json","events_json":"https://pith.science/api/pith-number/L6TZCUV5OYB5IU2FYWSKULNMAP/events.json","paper":"https://pith.science/paper/L6TZCUV5"},"agent_actions":{"view_html":"https://pith.science/pith/L6TZCUV5OYB5IU2FYWSKULNMAP","download_json":"https://pith.science/pith/L6TZCUV5OYB5IU2FYWSKULNMAP.json","view_paper":"https://pith.science/paper/L6TZCUV5","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2606.29350&json=true","fetch_graph":"https://pith.science/api/pith-number/L6TZCUV5OYB5IU2FYWSKULNMAP/graph.json","fetch_events":"https://pith.science/api/pith-number/L6TZCUV5OYB5IU2FYWSKULNMAP/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/L6TZCUV5OYB5IU2FYWSKULNMAP/action/timestamp_anchor","attest_storage":"https://pith.science/pith/L6TZCUV5OYB5IU2FYWSKULNMAP/action/storage_attestation","attest_author":"https://pith.science/pith/L6TZCUV5OYB5IU2FYWSKULNMAP/action/author_attestation","sign_citation":"https://pith.science/pith/L6TZCUV5OYB5IU2FYWSKULNMAP/action/citation_signature","submit_replication":"https://pith.science/pith/L6TZCUV5OYB5IU2FYWSKULNMAP/action/replication_record"}},"created_at":"2026-06-30T01:18:02.385823+00:00","updated_at":"2026-06-30T01:18:02.385823+00:00"}