{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:4CFZLSVTSTEM7KNWGIHRF4ANVG","short_pith_number":"pith:4CFZLSVT","schema_version":"1.0","canonical_sha256":"e08b95cab394c8cfa9b6320f12f00da990b2428d00d82b63bdc94b4541bc6d91","source":{"kind":"arxiv","id":"2607.17136","version":1},"attestation_state":"computed","paper":{"title":"Teach it to stop, not just to click","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.HC","cs.LG"],"primary_cat":"cs.SE","authors_text":"Barada Sahu (Cabal AI), Shivesh Pandey (Para AI)","submitted_at":"2026-07-19T08:46:06Z","abstract_excerpt":"Agentic computer-use RL is reported in single runs, and those numbers mislead. Using verifier-guided repair of a 35B computer-use agent (CUA) across five oracle-graded environments, we show a repaired policy's success rate is dominated by upstream variance: a variance-components decomposition across three cells (crossed data-draw $\\times$ seed grid, bootstrap CIs) finds evaluation variance negligible ($\\sigma_{\\mathrm{eval}} \\approx 0$) and the training-seed effect small everywhere ($\\leq 10\\%$); instead it splits between the data draw and run-to-run nondeterminism, the data draw's share risin"},"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":"2607.17136","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.SE","submitted_at":"2026-07-19T08:46:06Z","cross_cats_sorted":["cs.AI","cs.HC","cs.LG"],"title_canon_sha256":"c974215dbf0d323550f4a008e87939fecaf2ed25530d287f4841c0b613b5c427","abstract_canon_sha256":"d2c53124ee7981cb9b1ae55024eb1bce95c665a7449f4e73a4461c95c0a1207c"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-21T01:21:17.513388Z","signature_b64":"jc1beXLlQsLyWu6cVM421tr4x0O4fHtI4eBSCwkMEs7vk76d6B8uotXREJLyCuzSBWZIZyvdM8UDOavGR9EgDw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"e08b95cab394c8cfa9b6320f12f00da990b2428d00d82b63bdc94b4541bc6d91","last_reissued_at":"2026-07-21T01:21:17.512529Z","signature_status":"signed_v1","first_computed_at":"2026-07-21T01:21:17.512529Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Teach it to stop, not just to click","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.HC","cs.LG"],"primary_cat":"cs.SE","authors_text":"Barada Sahu (Cabal AI), Shivesh Pandey (Para AI)","submitted_at":"2026-07-19T08:46:06Z","abstract_excerpt":"Agentic computer-use RL is reported in single runs, and those numbers mislead. Using verifier-guided repair of a 35B computer-use agent (CUA) across five oracle-graded environments, we show a repaired policy's success rate is dominated by upstream variance: a variance-components decomposition across three cells (crossed data-draw $\\times$ seed grid, bootstrap CIs) finds evaluation variance negligible ($\\sigma_{\\mathrm{eval}} \\approx 0$) and the training-seed effect small everywhere ($\\leq 10\\%$); instead it splits between the data draw and run-to-run nondeterminism, the data draw's share risin"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2607.17136","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/2607.17136/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":"2607.17136","created_at":"2026-07-21T01:21:17.512962+00:00"},{"alias_kind":"arxiv_version","alias_value":"2607.17136v1","created_at":"2026-07-21T01:21:17.512962+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2607.17136","created_at":"2026-07-21T01:21:17.512962+00:00"},{"alias_kind":"pith_short_12","alias_value":"4CFZLSVTSTEM","created_at":"2026-07-21T01:21:17.512962+00:00"},{"alias_kind":"pith_short_16","alias_value":"4CFZLSVTSTEM7KNW","created_at":"2026-07-21T01:21:17.512962+00:00"},{"alias_kind":"pith_short_8","alias_value":"4CFZLSVT","created_at":"2026-07-21T01:21:17.512962+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/4CFZLSVTSTEM7KNWGIHRF4ANVG","json":"https://pith.science/pith/4CFZLSVTSTEM7KNWGIHRF4ANVG.json","graph_json":"https://pith.science/api/pith-number/4CFZLSVTSTEM7KNWGIHRF4ANVG/graph.json","events_json":"https://pith.science/api/pith-number/4CFZLSVTSTEM7KNWGIHRF4ANVG/events.json","paper":"https://pith.science/paper/4CFZLSVT"},"agent_actions":{"view_html":"https://pith.science/pith/4CFZLSVTSTEM7KNWGIHRF4ANVG","download_json":"https://pith.science/pith/4CFZLSVTSTEM7KNWGIHRF4ANVG.json","view_paper":"https://pith.science/paper/4CFZLSVT","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2607.17136&json=true","fetch_graph":"https://pith.science/api/pith-number/4CFZLSVTSTEM7KNWGIHRF4ANVG/graph.json","fetch_events":"https://pith.science/api/pith-number/4CFZLSVTSTEM7KNWGIHRF4ANVG/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/4CFZLSVTSTEM7KNWGIHRF4ANVG/action/timestamp_anchor","attest_storage":"https://pith.science/pith/4CFZLSVTSTEM7KNWGIHRF4ANVG/action/storage_attestation","attest_author":"https://pith.science/pith/4CFZLSVTSTEM7KNWGIHRF4ANVG/action/author_attestation","sign_citation":"https://pith.science/pith/4CFZLSVTSTEM7KNWGIHRF4ANVG/action/citation_signature","submit_replication":"https://pith.science/pith/4CFZLSVTSTEM7KNWGIHRF4ANVG/action/replication_record"}},"created_at":"2026-07-21T01:21:17.512962+00:00","updated_at":"2026-07-21T01:21:17.512962+00:00"}