{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2024:EUBKXFIYYKJWQFI6LBMN7HRI4I","short_pith_number":"pith:EUBKXFIY","canonical_record":{"source":{"id":"2410.07869","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2024-10-10T12:41:19Z","cross_cats_sorted":["cs.AI","cs.HC","cs.LG","cs.MA"],"title_canon_sha256":"516bd603bb4a0a3386c0a26dae17734a6b32c3168ab9979eb02dc322fba2ed56","abstract_canon_sha256":"5711da201dadf3ced8d75aca5700714f993b787a54608bca8c3ebbb3f2c210c1"},"schema_version":"1.0"},"canonical_sha256":"2502ab9518c29368151e5858df9e28e23e0e6cf82a474f4079ac20153ac2b602","source":{"kind":"arxiv","id":"2410.07869","version":3},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2410.07869","created_at":"2026-07-05T10:18:42Z"},{"alias_kind":"arxiv_version","alias_value":"2410.07869v3","created_at":"2026-07-05T10:18:42Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2410.07869","created_at":"2026-07-05T10:18:42Z"},{"alias_kind":"pith_short_12","alias_value":"EUBKXFIYYKJW","created_at":"2026-07-05T10:18:42Z"},{"alias_kind":"pith_short_16","alias_value":"EUBKXFIYYKJWQFI6","created_at":"2026-07-05T10:18:42Z"},{"alias_kind":"pith_short_8","alias_value":"EUBKXFIY","created_at":"2026-07-05T10:18:42Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2024:EUBKXFIYYKJWQFI6LBMN7HRI4I","target":"record","payload":{"canonical_record":{"source":{"id":"2410.07869","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2024-10-10T12:41:19Z","cross_cats_sorted":["cs.AI","cs.HC","cs.LG","cs.MA"],"title_canon_sha256":"516bd603bb4a0a3386c0a26dae17734a6b32c3168ab9979eb02dc322fba2ed56","abstract_canon_sha256":"5711da201dadf3ced8d75aca5700714f993b787a54608bca8c3ebbb3f2c210c1"},"schema_version":"1.0"},"canonical_sha256":"2502ab9518c29368151e5858df9e28e23e0e6cf82a474f4079ac20153ac2b602","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:18:42.893952Z","signature_b64":"dZX2n6nSZHbsXADOeEXf68rl1L0i4qOJHbShXe+KRD9FMgQjRh5+DAapsqLcQavDaSagL0xjont++qY+DK3GAA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"2502ab9518c29368151e5858df9e28e23e0e6cf82a474f4079ac20153ac2b602","last_reissued_at":"2026-07-05T10:18:42.893399Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:18:42.893399Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2410.07869","source_version":3,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T10:18:42Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"isPq97BaQYAJwMEfVQukjEkPqX/Gzn7+WZkjkomO4SbPNwHS/CHUqmrpGFBdXpBh1jmk4oDSL/I0puZ+wFRMCA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-17T18:53:53.298055Z"},"content_sha256":"85a8eccaf4c0a5e971244b46dc8e185be5da81b51ad45580c02f736cd3adfe71","schema_version":"1.0","event_id":"sha256:85a8eccaf4c0a5e971244b46dc8e185be5da81b51ad45580c02f736cd3adfe71"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2024:EUBKXFIYYKJWQFI6LBMN7HRI4I","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Benchmarking Agentic Workflow Generation","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.HC","cs.LG","cs.MA"],"primary_cat":"cs.CL","authors_text":"Fei Huang, Huajun Chen, Ningyu Zhang, Pengjun Xie, Runnan Fang, Shuofei Qiao, Xiaobin Wang, Yong Jiang, Zhisong Qiu","submitted_at":"2024-10-10T12:41:19Z","abstract_excerpt":"Large Language Models (LLMs), with their exceptional ability to handle a wide range of tasks, have driven significant advancements in tackling reasoning and planning tasks, wherein decomposing complex problems into executable workflows is a crucial step in this process. Existing workflow evaluation frameworks either focus solely on holistic performance or suffer from limitations such as restricted scenario coverage, simplistic workflow structures, and lax evaluation standards. To this end, we introduce WorfBench, a unified workflow generation benchmark with multi-faceted scenarios and intricat"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2410.07869","kind":"arxiv","version":3},"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/2410.07869/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"},"verdict_id":null},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T10:18:42Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"vu4uMQQXOXBoUcqgp57mlSSyQBFa2ee3oRlE6mpnKSPA1kMhoZUbHW5QZVPwLNs1TwByQ+FuuOHubvJriHnTDg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-17T18:53:53.298659Z"},"content_sha256":"906b1efbb38f7332645136c409860235098721201bb3eeaedb37619aeb79cb4a","schema_version":"1.0","event_id":"sha256:906b1efbb38f7332645136c409860235098721201bb3eeaedb37619aeb79cb4a"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/EUBKXFIYYKJWQFI6LBMN7HRI4I/bundle.json","state_url":"https://pith.science/pith/EUBKXFIYYKJWQFI6LBMN7HRI4I/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/EUBKXFIYYKJWQFI6LBMN7HRI4I/bundle.json","status":"primary"}],"public_keys":[{"key_id":"pith-v1-2026-05","algorithm":"ed25519","format":"raw","public_key_b64":"stVStoiQhXFxp4s2pdzPNoqVNBMojDU/fJ2db5S3CbM=","public_key_hex":"b2d552b68890857171a78b36a5dccf368a953413288c353f7c9d9d6f94b709b3","fingerprint_sha256_b32_first128bits":"RVFV5Z2OI2J3ZUO7ERDEBCYNKS","fingerprint_sha256_hex":"8d4b5ee74e4693bcd1df2446408b0d54","rotates_at":null,"url":"https://pith.science/pith-signing-key.json","notes":"Pith uses this Ed25519 key to sign canonical record SHA-256 digests. Verify with: ed25519_verify(public_key, message=canonical_sha256_bytes, signature=base64decode(signature_b64))."}],"merge_version":"pith-open-graph-merge-v1","built_at":"2026-08-17T18:53:53Z","links":{"resolver":"https://pith.science/pith/EUBKXFIYYKJWQFI6LBMN7HRI4I","bundle":"https://pith.science/pith/EUBKXFIYYKJWQFI6LBMN7HRI4I/bundle.json","state":"https://pith.science/pith/EUBKXFIYYKJWQFI6LBMN7HRI4I/state.json","well_known_bundle":"https://pith.science/.well-known/pith/EUBKXFIYYKJWQFI6LBMN7HRI4I/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:EUBKXFIYYKJWQFI6LBMN7HRI4I","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":"5711da201dadf3ced8d75aca5700714f993b787a54608bca8c3ebbb3f2c210c1","cross_cats_sorted":["cs.AI","cs.HC","cs.LG","cs.MA"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2024-10-10T12:41:19Z","title_canon_sha256":"516bd603bb4a0a3386c0a26dae17734a6b32c3168ab9979eb02dc322fba2ed56"},"schema_version":"1.0","source":{"id":"2410.07869","kind":"arxiv","version":3}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2410.07869","created_at":"2026-07-05T10:18:42Z"},{"alias_kind":"arxiv_version","alias_value":"2410.07869v3","created_at":"2026-07-05T10:18:42Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2410.07869","created_at":"2026-07-05T10:18:42Z"},{"alias_kind":"pith_short_12","alias_value":"EUBKXFIYYKJW","created_at":"2026-07-05T10:18:42Z"},{"alias_kind":"pith_short_16","alias_value":"EUBKXFIYYKJWQFI6","created_at":"2026-07-05T10:18:42Z"},{"alias_kind":"pith_short_8","alias_value":"EUBKXFIY","created_at":"2026-07-05T10:18:42Z"}],"graph_snapshots":[{"event_id":"sha256:906b1efbb38f7332645136c409860235098721201bb3eeaedb37619aeb79cb4a","target":"graph","created_at":"2026-07-05T10:18:42Z","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/2410.07869/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Large Language Models (LLMs), with their exceptional ability to handle a wide range of tasks, have driven significant advancements in tackling reasoning and planning tasks, wherein decomposing complex problems into executable workflows is a crucial step in this process. Existing workflow evaluation frameworks either focus solely on holistic performance or suffer from limitations such as restricted scenario coverage, simplistic workflow structures, and lax evaluation standards. To this end, we introduce WorfBench, a unified workflow generation benchmark with multi-faceted scenarios and intricat","authors_text":"Fei Huang, Huajun Chen, Ningyu Zhang, Pengjun Xie, Runnan Fang, Shuofei Qiao, Xiaobin Wang, Yong Jiang, Zhisong Qiu","cross_cats":["cs.AI","cs.HC","cs.LG","cs.MA"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2024-10-10T12:41:19Z","title":"Benchmarking Agentic Workflow Generation"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2410.07869","kind":"arxiv","version":3},"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":"sha256:85a8eccaf4c0a5e971244b46dc8e185be5da81b51ad45580c02f736cd3adfe71","target":"record","created_at":"2026-07-05T10:18:42Z","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":"5711da201dadf3ced8d75aca5700714f993b787a54608bca8c3ebbb3f2c210c1","cross_cats_sorted":["cs.AI","cs.HC","cs.LG","cs.MA"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2024-10-10T12:41:19Z","title_canon_sha256":"516bd603bb4a0a3386c0a26dae17734a6b32c3168ab9979eb02dc322fba2ed56"},"schema_version":"1.0","source":{"id":"2410.07869","kind":"arxiv","version":3}},"canonical_sha256":"2502ab9518c29368151e5858df9e28e23e0e6cf82a474f4079ac20153ac2b602","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"2502ab9518c29368151e5858df9e28e23e0e6cf82a474f4079ac20153ac2b602","first_computed_at":"2026-07-05T10:18:42.893399Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T10:18:42.893399Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"dZX2n6nSZHbsXADOeEXf68rl1L0i4qOJHbShXe+KRD9FMgQjRh5+DAapsqLcQavDaSagL0xjont++qY+DK3GAA==","signature_status":"signed_v1","signed_at":"2026-07-05T10:18:42.893952Z","signed_message":"canonical_sha256_bytes"},"source_id":"2410.07869","source_kind":"arxiv","source_version":3}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:85a8eccaf4c0a5e971244b46dc8e185be5da81b51ad45580c02f736cd3adfe71","sha256:906b1efbb38f7332645136c409860235098721201bb3eeaedb37619aeb79cb4a"],"state_sha256":"af70f7a755be59fc935b36c476c63c31147e976b762fcc3b8e01da20c8f647b2"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"irBbFHTOPHhJiz5tIL/ro9KGbcVmrkVdmM/4/B8oNTyRBaZS99+UcjfxkrARoX/QsqTtAMS9gW73FLpSUWWyDg==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-17T18:53:53.305507Z","bundle_sha256":"11d7a0f20be86383a44d2afc1bc97ea2e3c128e5e1d04a3e98732ed06b2478a1"}}