{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2023:OSAQPBI6OK7GIN2XTKUCBEESUI","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":"6c53eec99c54230a8983da974f32025c5828cf8cc3c14dca3c34f901f2a4d9c4","cross_cats_sorted":["cs.AI"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2023-11-30T18:02:44Z","title_canon_sha256":"286a6e8fac6874174a0790aafc029380ddb31e45a88da4363a225aff9d63e070"},"schema_version":"1.0","source":{"id":"2311.18760","kind":"arxiv","version":4}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2311.18760","created_at":"2026-07-05T09:29:24Z"},{"alias_kind":"arxiv_version","alias_value":"2311.18760v4","created_at":"2026-07-05T09:29:24Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2311.18760","created_at":"2026-07-05T09:29:24Z"},{"alias_kind":"pith_short_12","alias_value":"OSAQPBI6OK7G","created_at":"2026-07-05T09:29:24Z"},{"alias_kind":"pith_short_16","alias_value":"OSAQPBI6OK7GIN2X","created_at":"2026-07-05T09:29:24Z"},{"alias_kind":"pith_short_8","alias_value":"OSAQPBI6","created_at":"2026-07-05T09:29:24Z"}],"graph_snapshots":[{"event_id":"sha256:93a940f3fc37a4c664d1e9c0a9b0a0b86af06a2d6c93a84ebabe0bbf90f61c9a","target":"graph","created_at":"2026-07-05T09:29:24Z","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/2311.18760/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"In recent years, the remarkable progress of large language models (LLMs) has sparked interest in task automation, which involves decomposing complex tasks described by user instructions into sub-tasks and invoking external tools to execute them, playing a central role in autonomous agents. However, there is a lack of systematic and standardized benchmarks to promote the development of LLMs in task automation. To address this, we introduce TaskBench, a comprehensive framework to evaluate the capability of LLMs in task automation. Specifically, task automation can be divided into three critical ","authors_text":"Dongsheng Li, Kaitao Song, Kan Ren, Siyu Yuan, Weiming Lu, Wenqi Zhang, Xu Tan, Yongliang Shen, Yueting Zhuang","cross_cats":["cs.AI"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2023-11-30T18:02:44Z","title":"TaskBench: Benchmarking Large Language Models for Task Automation"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2311.18760","kind":"arxiv","version":4},"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":"sha256:6d6ef787ac27a8a8377e370194a9189021e129c747b86aa8528afa4150f00fee","target":"record","created_at":"2026-07-05T09:29:24Z","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":"6c53eec99c54230a8983da974f32025c5828cf8cc3c14dca3c34f901f2a4d9c4","cross_cats_sorted":["cs.AI"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2023-11-30T18:02:44Z","title_canon_sha256":"286a6e8fac6874174a0790aafc029380ddb31e45a88da4363a225aff9d63e070"},"schema_version":"1.0","source":{"id":"2311.18760","kind":"arxiv","version":4}},"canonical_sha256":"748107851e72be6437579aa8209092a21f9d15a1d07a1860ed2eaad77c372923","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"748107851e72be6437579aa8209092a21f9d15a1d07a1860ed2eaad77c372923","first_computed_at":"2026-07-05T09:29:24.916468Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T09:29:24.916468Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"pw5KQ2xm/MeS3njw7Gdp/b2qVO7vjRs7rDzMYp/7b2bv3osJA5shJj+rr1bwFN0zafb0maUG0paMvycX0VpsBQ==","signature_status":"signed_v1","signed_at":"2026-07-05T09:29:24.916962Z","signed_message":"canonical_sha256_bytes"},"source_id":"2311.18760","source_kind":"arxiv","source_version":4}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:6d6ef787ac27a8a8377e370194a9189021e129c747b86aa8528afa4150f00fee","sha256:93a940f3fc37a4c664d1e9c0a9b0a0b86af06a2d6c93a84ebabe0bbf90f61c9a"],"state_sha256":"8c036348fb1f380ef1ac4e671c2c4af816e54025a2b353609807c08311f61240"}