{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:ENLSPLV2VBZC5Z2CYJXCFHK4YZ","merge_version":"pith-open-graph-merge-v1","event_count":2,"valid_event_count":2,"invalid_event_count":0,"equivocation_count":0,"current":{"canonical_record":{"metadata":{"abstract_canon_sha256":"56f08b56ac11b34970ff0149b743534d398ca842719913534b1585f4fbeb268f","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-10-10T21:20:51Z","title_canon_sha256":"bb3790d6f7234d3504fac410e90dc27d7e0fab9e79c48dc77ac1d6f3632db177"},"schema_version":"1.0","source":{"id":"2510.09872","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2510.09872","created_at":"2026-05-20T01:04:59Z"},{"alias_kind":"arxiv_version","alias_value":"2510.09872v2","created_at":"2026-05-20T01:04:59Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2510.09872","created_at":"2026-05-20T01:04:59Z"},{"alias_kind":"pith_short_12","alias_value":"ENLSPLV2VBZC","created_at":"2026-05-20T01:04:59Z"},{"alias_kind":"pith_short_16","alias_value":"ENLSPLV2VBZC5Z2C","created_at":"2026-05-20T01:04:59Z"},{"alias_kind":"pith_short_8","alias_value":"ENLSPLV2","created_at":"2026-05-20T01:04:59Z"}],"graph_snapshots":[{"event_id":"sha256:357a9860122fa4a9254df80b5313b62b29604e8f375d3d09c68a2a458ef5fbc7","target":"graph","created_at":"2026-05-20T01:04:59Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"graph_snapshot":{"author_claims":{"count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","strong_count":0},"builder_version":"pith-number-builder-2026-05-17-v1","claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"integrity":{"available":true,"clean":true,"detectors_run":[],"endpoint":"/pith/2510.09872/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Training web agents to navigate complex, real-world websites requires them to master $\\textit{subtasks}$ - short-horizon interactions on multiple UI components (e.g., choosing the correct date in a date picker, or scrolling in a container to extract information). We introduce WARC-Bench (Web Archive Benchmark), a novel web navigation benchmark featuring 438 tasks designed to evaluate multimodal AI agents on subtasks. WARC-Bench enables sandboxed interactions with dynamic and realistic webpages using Web ARChive files. We show that WARC-Bench is challenging for leading computer-use models, with","authors_text":"Charlene Y. Lee, Cheng Chang, Gang Li, Ignacio Cases, Manpreet Kaur, Peng Qi, Rishu Garg, Sanjari Srivastava, Yanan Xie, Yining Mao, Yuezhang Li","cross_cats":["cs.AI"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-10-10T21:20:51Z","title":"WARC-Bench: Web Archive Based Benchmark for GUI Subtask Executions"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2510.09872","kind":"arxiv","version":2},"verdict":{"created_at":null,"id":null,"model_set":{},"one_line_summary":"","pipeline_version":null,"pith_extraction_headline":"","strongest_claim":"","weakest_assumption":""}},"verdict_id":null}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:f51b77785412d5e615409cdd0181febe39f53498ba9ad7a012c6f147359408cc","target":"record","created_at":"2026-05-20T01:04:59Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"attestation_state":"computed","canonical_record":{"metadata":{"abstract_canon_sha256":"56f08b56ac11b34970ff0149b743534d398ca842719913534b1585f4fbeb268f","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-10-10T21:20:51Z","title_canon_sha256":"bb3790d6f7234d3504fac410e90dc27d7e0fab9e79c48dc77ac1d6f3632db177"},"schema_version":"1.0","source":{"id":"2510.09872","kind":"arxiv","version":2}},"canonical_sha256":"235727aebaa8722ee742c26e229d5cc66e0198f436d5fb81fa73ac8e5a2e6200","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"235727aebaa8722ee742c26e229d5cc66e0198f436d5fb81fa73ac8e5a2e6200","first_computed_at":"2026-05-20T01:04:59.903716Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-20T01:04:59.903716Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"Zy5NiN+e1qVT3ehJQ5pK3VHn5DETdQnXqQd3SiQ7IpVAcgwcDyTWVXka3w15y47STtU/mBbZExl1mGASvZGDAg==","signature_status":"signed_v1","signed_at":"2026-05-20T01:04:59.904722Z","signed_message":"canonical_sha256_bytes"},"source_id":"2510.09872","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:f51b77785412d5e615409cdd0181febe39f53498ba9ad7a012c6f147359408cc","sha256:357a9860122fa4a9254df80b5313b62b29604e8f375d3d09c68a2a458ef5fbc7"],"state_sha256":"a8ec7d55c23680c507f3d788aa4e9a1d7ba24a65f0c830bd8b51c2c5cf990b4d"}