{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:3KECOAG27EKT7EVDMBOTHH3LU3","short_pith_number":"pith:3KECOAG2","schema_version":"1.0","canonical_sha256":"da882700daf9153f92a3605d339f6ba6ddbabefb3cda0ee048296a78b3bbf927","source":{"kind":"arxiv","id":"2605.11817","version":2},"attestation_state":"computed","paper":{"title":"See What Matters: Differentiable Grid Sample Pruning for Generalizable Vision-Language-Action Model","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"Differentiable grid sampling lets VLA models compress visual tokens to under 10 percent while keeping full manipulation success.","cross_cats":["cs.CV"],"primary_cat":"cs.RO","authors_text":"Chang Xu, Chengbin Du, Chenghao Xia, Yanxiang Ma, Yixu Feng, Yunke Wang, Zinan Zhao","submitted_at":"2026-05-12T09:08:42Z","abstract_excerpt":"Vision-Language-Action (VLA) models have shown remarkable promise in robotics manipulation, yet their high computational cost hinders real-time deployment. Existing token pruning methods suffer from a fundamental trade-off: aggressive compression using pruning inevitably discards critical geometric details like contact points, leading to severe performance degradation. This forces a compromise, limiting the achievable compression rate and thus the potential speedup. We argue that breaking this trade-off requires rethinking compression as a geometry-aware, continuous token resampling in the vis"},"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":true},"canonical_record":{"source":{"id":"2605.11817","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.RO","submitted_at":"2026-05-12T09:08:42Z","cross_cats_sorted":["cs.CV"],"title_canon_sha256":"d890d3fb46f5b83e15da28afa517f3d7a715d5bb097cd3daccd743f67ac7048e","abstract_canon_sha256":"10064b6bada4125ae37b0039636c0aaf75c085318aebb5df5037928af974560d"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-20T00:04:36.179164Z","signature_b64":"qspYdYi+6kwO7am0kfFS9StvP19hj4j+aXBEgZ6zOlGHnFZ0kCC7o/Q2Y5ivQdqzZpuWggayqgbQp4d9Q/gkDw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"da882700daf9153f92a3605d339f6ba6ddbabefb3cda0ee048296a78b3bbf927","last_reissued_at":"2026-05-20T00:04:36.178020Z","signature_status":"signed_v1","first_computed_at":"2026-05-20T00:04:36.178020Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"See What Matters: Differentiable Grid Sample Pruning for Generalizable Vision-Language-Action Model","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"Differentiable grid sampling lets VLA models compress visual tokens to under 10 percent while keeping full manipulation success.","cross_cats":["cs.CV"],"primary_cat":"cs.RO","authors_text":"Chang Xu, Chengbin Du, Chenghao Xia, Yanxiang Ma, Yixu Feng, Yunke Wang, Zinan Zhao","submitted_at":"2026-05-12T09:08:42Z","abstract_excerpt":"Vision-Language-Action (VLA) models have shown remarkable promise in robotics manipulation, yet their high computational cost hinders real-time deployment. Existing token pruning methods suffer from a fundamental trade-off: aggressive compression using pruning inevitably discards critical geometric details like contact points, leading to severe performance degradation. This forces a compromise, limiting the achievable compression rate and thus the potential speedup. We argue that breaking this trade-off requires rethinking compression as a geometry-aware, continuous token resampling in the vis"},"claims":{"count":4,"items":[{"kind":"strongest_claim","text":"validating the lowest feasible visual token count reported to date, GridS achieves a 76% reduction in FLOPs with no degradation in the success rate.","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","text":"That adaptively predicting a minimal set of salient coordinates and extracting features via differentiable interpolation will always preserve the critical geometric details (contact points, spatial relations) required for manipulation success across tasks and environments.","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","text":"GridS reduces visual tokens in VLA models to under 10% of the original count via task-aware differentiable resampling, delivering 76% lower FLOPs with no drop in task success rate on benchmarks and real robots.","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"},{"kind":"headline","text":"Differentiable grid sampling lets VLA models compress visual tokens to under 10 percent while keeping full manipulation success.","source":"verdict.pith_extraction.headline","status":"machine_extracted","claim_id":"C4","attestation":"unclaimed"}],"snapshot_sha256":"b7a26ca9e662d38a871a961c34d9c9af25335b81066568a1e799cdd4343096e4"},"source":{"id":"2605.11817","kind":"arxiv","version":2},"verdict":{"id":"02473b66-5ced-4ed6-aa02-292631ff9758","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-13T05:37:23.048495Z","strongest_claim":"validating the lowest feasible visual token count reported to date, GridS achieves a 76% reduction in FLOPs with no degradation in the success rate.","one_line_summary":"GridS reduces visual tokens in VLA models to under 10% of the original count via task-aware differentiable resampling, delivering 76% lower FLOPs with no drop in task success rate on benchmarks and real robots.","pipeline_version":"pith-pipeline@v0.9.0","weakest_assumption":"That adaptively predicting a minimal set of salient coordinates and extracting features via differentiable interpolation will always preserve the critical geometric details (contact points, spatial relations) required for manipulation success across tasks and environments.","pith_extraction_headline":"Differentiable grid sampling lets VLA models compress visual tokens to under 10 percent while keeping full manipulation success."},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2605.11817/integrity.json","findings":[],"available":true,"detectors_run":[{"name":"ai_meta_artifact","ran_at":"2026-05-19T11:37:09.344165Z","status":"completed","version":"1.0.0","findings_count":0},{"name":"doi_title_agreement","ran_at":"2026-05-19T09:01:18.728167Z","status":"completed","version":"1.0.0","findings_count":0},{"name":"doi_compliance","ran_at":"2026-05-19T08:09:33.258817Z","status":"completed","version":"1.0.0","findings_count":0}],"snapshot_sha256":"5369e7ac0214a6872f8302180b02f9bfb6ef7fedf5ac01a1556e47fe9a8a1d7f"},"references":{"count":0,"sample":[],"resolved_work":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","internal_anchors":0},"formal_canon":{"evidence_count":1,"snapshot_sha256":"bc3aa9c3266c2b61e4fe9b2fd3c9ad46fd5fe3dd562c25bf4020765d7a7ffc98"},"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":"2605.11817","created_at":"2026-05-20T00:04:36.178195+00:00"},{"alias_kind":"arxiv_version","alias_value":"2605.11817v2","created_at":"2026-05-20T00:04:36.178195+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.11817","created_at":"2026-05-20T00:04:36.178195+00:00"},{"alias_kind":"pith_short_12","alias_value":"3KECOAG27EKT","created_at":"2026-05-20T00:04:36.178195+00:00"},{"alias_kind":"pith_short_16","alias_value":"3KECOAG27EKT7EVD","created_at":"2026-05-20T00:04:36.178195+00:00"},{"alias_kind":"pith_short_8","alias_value":"3KECOAG2","created_at":"2026-05-20T00:04:36.178195+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":1,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/3KECOAG27EKT7EVDMBOTHH3LU3","json":"https://pith.science/pith/3KECOAG27EKT7EVDMBOTHH3LU3.json","graph_json":"https://pith.science/api/pith-number/3KECOAG27EKT7EVDMBOTHH3LU3/graph.json","events_json":"https://pith.science/api/pith-number/3KECOAG27EKT7EVDMBOTHH3LU3/events.json","paper":"https://pith.science/paper/3KECOAG2"},"agent_actions":{"view_html":"https://pith.science/pith/3KECOAG27EKT7EVDMBOTHH3LU3","download_json":"https://pith.science/pith/3KECOAG27EKT7EVDMBOTHH3LU3.json","view_paper":"https://pith.science/paper/3KECOAG2","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2605.11817&json=true","fetch_graph":"https://pith.science/api/pith-number/3KECOAG27EKT7EVDMBOTHH3LU3/graph.json","fetch_events":"https://pith.science/api/pith-number/3KECOAG27EKT7EVDMBOTHH3LU3/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/3KECOAG27EKT7EVDMBOTHH3LU3/action/timestamp_anchor","attest_storage":"https://pith.science/pith/3KECOAG27EKT7EVDMBOTHH3LU3/action/storage_attestation","attest_author":"https://pith.science/pith/3KECOAG27EKT7EVDMBOTHH3LU3/action/author_attestation","sign_citation":"https://pith.science/pith/3KECOAG27EKT7EVDMBOTHH3LU3/action/citation_signature","submit_replication":"https://pith.science/pith/3KECOAG27EKT7EVDMBOTHH3LU3/action/replication_record"}},"created_at":"2026-05-20T00:04:36.178195+00:00","updated_at":"2026-05-20T00:04:36.178195+00:00"}