{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:RVL3OSTAMO5G5Z3MJTSNS3UDPJ","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":"0cafd2c79559a89480ad6ea2fe47f764015c653be83b152c16acb4a165cfe0f0","cross_cats_sorted":["cs.AI","cs.DB"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-08-15T00:58:10Z","title_canon_sha256":"585bd99f0b945af1348d4df0a7e3673a39e388844d77b13b316d8f17a91facdd"},"schema_version":"1.0","source":{"id":"2508.11133","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2508.11133","created_at":"2026-07-05T12:04:17Z"},{"alias_kind":"arxiv_version","alias_value":"2508.11133v2","created_at":"2026-07-05T12:04:17Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2508.11133","created_at":"2026-07-05T12:04:17Z"},{"alias_kind":"pith_short_12","alias_value":"RVL3OSTAMO5G","created_at":"2026-07-05T12:04:17Z"},{"alias_kind":"pith_short_16","alias_value":"RVL3OSTAMO5G5Z3M","created_at":"2026-07-05T12:04:17Z"},{"alias_kind":"pith_short_8","alias_value":"RVL3OSTA","created_at":"2026-07-05T12:04:17Z"}],"graph_snapshots":[{"event_id":"sha256:8685f738dd4201c9ee8eec38b1d0be294883ac8f30e95ce0d935ea4531bb36da","target":"graph","created_at":"2026-07-05T12:04:17Z","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/2508.11133/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Automated agents, powered by Large language models (LLMs), are emerging as the go-to tool for querying information. However, evaluation benchmarks for LLM agents rarely feature natural questions that are both information-seeking and genuinely time-consuming for humans. To address this gap we introduce MoNaCo, a benchmark of 1,315 natural and time-consuming questions that require dozens, and at times hundreds, of intermediate steps to solve -- far more than any existing QA benchmark. To build MoNaCo, we developed a decomposed annotation pipeline to elicit and manually answer real-world time-con","authors_text":"Ashish Sabharwal, Dan Roth, Harsh Trivedi, Mor Geva, Reut Tsarfaty, Tomer Wolfson, Tushar Khot, Yoav Goldberg","cross_cats":["cs.AI","cs.DB"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-08-15T00:58:10Z","title":"MoNaCo: More Natural and Complex Questions for Reasoning Across Dozens of Documents"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2508.11133","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:f52089e57e6d4b9799fb0541f215bcbc4d362d9bdf20423a9409c69c1a4b3394","target":"record","created_at":"2026-07-05T12:04:17Z","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":"0cafd2c79559a89480ad6ea2fe47f764015c653be83b152c16acb4a165cfe0f0","cross_cats_sorted":["cs.AI","cs.DB"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-08-15T00:58:10Z","title_canon_sha256":"585bd99f0b945af1348d4df0a7e3673a39e388844d77b13b316d8f17a91facdd"},"schema_version":"1.0","source":{"id":"2508.11133","kind":"arxiv","version":2}},"canonical_sha256":"8d57b74a6063ba6ee76c4ce4d96e837a6b9e72bbf0db1ada5a92ddf84c5f9aae","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"8d57b74a6063ba6ee76c4ce4d96e837a6b9e72bbf0db1ada5a92ddf84c5f9aae","first_computed_at":"2026-07-05T12:04:17.099720Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T12:04:17.099720Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"xSqqh83C3AOTYlv5xNrTbK8WQi6DvUOTb6EgIz5SdlLaFTvwHbNdKEtBWGstZIVeE7+0rAqQMFtVsHesiDdaCw==","signature_status":"signed_v1","signed_at":"2026-07-05T12:04:17.100262Z","signed_message":"canonical_sha256_bytes"},"source_id":"2508.11133","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:f52089e57e6d4b9799fb0541f215bcbc4d362d9bdf20423a9409c69c1a4b3394","sha256:8685f738dd4201c9ee8eec38b1d0be294883ac8f30e95ce0d935ea4531bb36da"],"state_sha256":"624aa26b172755526495ac024008128eb454c0449b9e24fe6bbe6631837cdb80"}