{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:4QWADUUYJ6NI4EOIXM455C62OV","short_pith_number":"pith:4QWADUUY","schema_version":"1.0","canonical_sha256":"e42c01d2984f9a8e11c8bb39de8bda75423270f1155a1d84c618f67717262a2d","source":{"kind":"arxiv","id":"2606.20724","version":1},"attestation_state":"computed","paper":{"title":"When Web Agents Finish but Still Fail: Reproducible Triggers and Trace Diagnostics for Parallel Web Exploration","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.AI","authors_text":"Aagam Sogani, Botao Rui, Minghao Yan, Rishi Agarwal, Shivaram Venkataraman, Swetha Vaidyanathan","submitted_at":"2026-06-16T23:00:25Z","abstract_excerpt":"Long-horizon web agents often fail in ways hidden by final-answer evaluation: they may visit useful pages, produce a well-formed answer, and terminate confidently while still missing fields, over-including unsupported items, or relying on stale evidence. We study these failures with Parallel WebBench, a parallel web-exploration benchmark containing 1,679 verified records: 350 manually curated parallel tasks and 1,329 reconstructed records with verified URL-based trajectories. We train WebExplorer-style agents with GRPO under human-only, balanced human-synthetic, and synthetic-heavy data mixtur"},"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":"2606.20724","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2026-06-16T23:00:25Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"9f0e6112b81a92ce10c647589e230bb1fa0a02a4875b81690ee3c70e0fc051c5","abstract_canon_sha256":"f42f044894a5468f055ee847349035f4011f6edabc8c179634d5a67ff7b1220c"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-23T00:11:54.764153Z","signature_b64":"JZESq2u4msqEmYXADmJGF8wD+/EEozmtio4qGlyNYRjKWeZ2/NB/6jUELa1bHE5MQicPRwAq3D8DPrh8ol1BAA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"e42c01d2984f9a8e11c8bb39de8bda75423270f1155a1d84c618f67717262a2d","last_reissued_at":"2026-06-23T00:11:54.763708Z","signature_status":"signed_v1","first_computed_at":"2026-06-23T00:11:54.763708Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"When Web Agents Finish but Still Fail: Reproducible Triggers and Trace Diagnostics for Parallel Web Exploration","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.AI","authors_text":"Aagam Sogani, Botao Rui, Minghao Yan, Rishi Agarwal, Shivaram Venkataraman, Swetha Vaidyanathan","submitted_at":"2026-06-16T23:00:25Z","abstract_excerpt":"Long-horizon web agents often fail in ways hidden by final-answer evaluation: they may visit useful pages, produce a well-formed answer, and terminate confidently while still missing fields, over-including unsupported items, or relying on stale evidence. We study these failures with Parallel WebBench, a parallel web-exploration benchmark containing 1,679 verified records: 350 manually curated parallel tasks and 1,329 reconstructed records with verified URL-based trajectories. We train WebExplorer-style agents with GRPO under human-only, balanced human-synthetic, and synthetic-heavy data mixtur"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2606.20724","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/2606.20724/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":"2606.20724","created_at":"2026-06-23T00:11:54.763790+00:00"},{"alias_kind":"arxiv_version","alias_value":"2606.20724v1","created_at":"2026-06-23T00:11:54.763790+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2606.20724","created_at":"2026-06-23T00:11:54.763790+00:00"},{"alias_kind":"pith_short_12","alias_value":"4QWADUUYJ6NI","created_at":"2026-06-23T00:11:54.763790+00:00"},{"alias_kind":"pith_short_16","alias_value":"4QWADUUYJ6NI4EOI","created_at":"2026-06-23T00:11:54.763790+00:00"},{"alias_kind":"pith_short_8","alias_value":"4QWADUUY","created_at":"2026-06-23T00:11:54.763790+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/4QWADUUYJ6NI4EOIXM455C62OV","json":"https://pith.science/pith/4QWADUUYJ6NI4EOIXM455C62OV.json","graph_json":"https://pith.science/api/pith-number/4QWADUUYJ6NI4EOIXM455C62OV/graph.json","events_json":"https://pith.science/api/pith-number/4QWADUUYJ6NI4EOIXM455C62OV/events.json","paper":"https://pith.science/paper/4QWADUUY"},"agent_actions":{"view_html":"https://pith.science/pith/4QWADUUYJ6NI4EOIXM455C62OV","download_json":"https://pith.science/pith/4QWADUUYJ6NI4EOIXM455C62OV.json","view_paper":"https://pith.science/paper/4QWADUUY","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2606.20724&json=true","fetch_graph":"https://pith.science/api/pith-number/4QWADUUYJ6NI4EOIXM455C62OV/graph.json","fetch_events":"https://pith.science/api/pith-number/4QWADUUYJ6NI4EOIXM455C62OV/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/4QWADUUYJ6NI4EOIXM455C62OV/action/timestamp_anchor","attest_storage":"https://pith.science/pith/4QWADUUYJ6NI4EOIXM455C62OV/action/storage_attestation","attest_author":"https://pith.science/pith/4QWADUUYJ6NI4EOIXM455C62OV/action/author_attestation","sign_citation":"https://pith.science/pith/4QWADUUYJ6NI4EOIXM455C62OV/action/citation_signature","submit_replication":"https://pith.science/pith/4QWADUUYJ6NI4EOIXM455C62OV/action/replication_record"}},"created_at":"2026-06-23T00:11:54.763790+00:00","updated_at":"2026-06-23T00:11:54.763790+00:00"}