{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:SF55K56K4OHQSKXWED3FVGDMXM","short_pith_number":"pith:SF55K56K","schema_version":"1.0","canonical_sha256":"917bd577cae38f092af620f65a986cbb2219ab069752aff8e35d4deadf27dc7d","source":{"kind":"arxiv","id":"2605.16154","version":1},"attestation_state":"computed","paper":{"title":"Learn Where Outcomes Diverge: Efficient VLA RL via Probabilistic Chunk Masking","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.RO"],"primary_cat":"cs.LG","authors_text":"Nikshep Grampurohit, Pulkit Verma, Vaidehi Bagaria","submitted_at":"2026-05-15T16:33:59Z","abstract_excerpt":"Reinforcement learning (RL) allows vision-language-action (VLA) policies to generalize beyond their training distribution by optimizing directly for task success, but post-training is computationally expensive. A natural response has been to speed rollout collection through faster simulators and world models. In GRPO-based VLA RL, we find that the dominant cost lies elsewhere: gradient computation accounts for approximately 78% of wall-clock time per step in our runs, while rollout collection accounts for only 21%. Gradient cost dominates because much of this computation is spent on phases tha"},"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":"2605.16154","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-05-15T16:33:59Z","cross_cats_sorted":["cs.RO"],"title_canon_sha256":"664582dafb2bb99f24dca92613a7ab8cc1a97fddf2ae6194f98d821d1021b944","abstract_canon_sha256":"f234a20cc377ead2bcfd6d6544e73879b9a5f7b1f7cec32f2b739fd94270483f"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-20T00:01:55.311866Z","signature_b64":"ZogY617mHY0pl8IqKN8w3uC5zx0IkjkRDq8v2CPpMALb+tvKQmt0FAUWw7dp8jHYBcBvoILSMOlJvHAFX57fCA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"917bd577cae38f092af620f65a986cbb2219ab069752aff8e35d4deadf27dc7d","last_reissued_at":"2026-05-20T00:01:55.310797Z","signature_status":"signed_v1","first_computed_at":"2026-05-20T00:01:55.310797Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Learn Where Outcomes Diverge: Efficient VLA RL via Probabilistic Chunk Masking","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.RO"],"primary_cat":"cs.LG","authors_text":"Nikshep Grampurohit, Pulkit Verma, Vaidehi Bagaria","submitted_at":"2026-05-15T16:33:59Z","abstract_excerpt":"Reinforcement learning (RL) allows vision-language-action (VLA) policies to generalize beyond their training distribution by optimizing directly for task success, but post-training is computationally expensive. A natural response has been to speed rollout collection through faster simulators and world models. In GRPO-based VLA RL, we find that the dominant cost lies elsewhere: gradient computation accounts for approximately 78% of wall-clock time per step in our runs, while rollout collection accounts for only 21%. Gradient cost dominates because much of this computation is spent on phases tha"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2605.16154","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/2605.16154/integrity.json","findings":[],"available":true,"detectors_run":[{"name":"cited_work_retraction","ran_at":"2026-05-19T18:22:02.507774Z","status":"completed","version":"1.0.0","findings_count":0},{"name":"ai_meta_artifact","ran_at":"2026-05-19T17:33:31.040869Z","status":"skipped","version":"1.0.0","findings_count":0},{"name":"claim_evidence","ran_at":"2026-05-19T16:41:55.442418Z","status":"completed","version":"1.0.0","findings_count":0}],"snapshot_sha256":"0870b263bfddb3599c0d92368e067d8ad8d3a764319d73445a6cfc5d40bfde6a"},"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":"2605.16154","created_at":"2026-05-20T00:01:55.310986+00:00"},{"alias_kind":"arxiv_version","alias_value":"2605.16154v1","created_at":"2026-05-20T00:01:55.310986+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.16154","created_at":"2026-05-20T00:01:55.310986+00:00"},{"alias_kind":"pith_short_12","alias_value":"SF55K56K4OHQ","created_at":"2026-05-20T00:01:55.310986+00:00"},{"alias_kind":"pith_short_16","alias_value":"SF55K56K4OHQSKXW","created_at":"2026-05-20T00:01:55.310986+00:00"},{"alias_kind":"pith_short_8","alias_value":"SF55K56K","created_at":"2026-05-20T00:01:55.310986+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/SF55K56K4OHQSKXWED3FVGDMXM","json":"https://pith.science/pith/SF55K56K4OHQSKXWED3FVGDMXM.json","graph_json":"https://pith.science/api/pith-number/SF55K56K4OHQSKXWED3FVGDMXM/graph.json","events_json":"https://pith.science/api/pith-number/SF55K56K4OHQSKXWED3FVGDMXM/events.json","paper":"https://pith.science/paper/SF55K56K"},"agent_actions":{"view_html":"https://pith.science/pith/SF55K56K4OHQSKXWED3FVGDMXM","download_json":"https://pith.science/pith/SF55K56K4OHQSKXWED3FVGDMXM.json","view_paper":"https://pith.science/paper/SF55K56K","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2605.16154&json=true","fetch_graph":"https://pith.science/api/pith-number/SF55K56K4OHQSKXWED3FVGDMXM/graph.json","fetch_events":"https://pith.science/api/pith-number/SF55K56K4OHQSKXWED3FVGDMXM/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/SF55K56K4OHQSKXWED3FVGDMXM/action/timestamp_anchor","attest_storage":"https://pith.science/pith/SF55K56K4OHQSKXWED3FVGDMXM/action/storage_attestation","attest_author":"https://pith.science/pith/SF55K56K4OHQSKXWED3FVGDMXM/action/author_attestation","sign_citation":"https://pith.science/pith/SF55K56K4OHQSKXWED3FVGDMXM/action/citation_signature","submit_replication":"https://pith.science/pith/SF55K56K4OHQSKXWED3FVGDMXM/action/replication_record"}},"created_at":"2026-05-20T00:01:55.310986+00:00","updated_at":"2026-05-20T00:01:55.310986+00:00"}