{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2024:6VCT5D7WARX4OT4PX3I6GG64RR","short_pith_number":"pith:6VCT5D7W","canonical_record":{"source":{"id":"2405.00823","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-05-01T19:07:03Z","cross_cats_sorted":["cs.AI","cs.MA"],"title_canon_sha256":"fc5600aec32d7565a760f8311e55f6712cb479366969b2fb3ba8c5d7a453736c","abstract_canon_sha256":"f1210784f75fa632740eb133c47a4642fae787aa2623e4f4d1a3d8b3af46de87"},"schema_version":"1.0"},"canonical_sha256":"f5453e8ff6046fc74f8fbed1e31bdc8c47f1716a5f3c0dcba4f07dae8ff81879","source":{"kind":"arxiv","id":"2405.00823","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2405.00823","created_at":"2026-07-05T08:51:51Z"},{"alias_kind":"arxiv_version","alias_value":"2405.00823v2","created_at":"2026-07-05T08:51:51Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2405.00823","created_at":"2026-07-05T08:51:51Z"},{"alias_kind":"pith_short_12","alias_value":"6VCT5D7WARX4","created_at":"2026-07-05T08:51:51Z"},{"alias_kind":"pith_short_16","alias_value":"6VCT5D7WARX4OT4P","created_at":"2026-07-05T08:51:51Z"},{"alias_kind":"pith_short_8","alias_value":"6VCT5D7W","created_at":"2026-07-05T08:51:51Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2024:6VCT5D7WARX4OT4PX3I6GG64RR","target":"record","payload":{"canonical_record":{"source":{"id":"2405.00823","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-05-01T19:07:03Z","cross_cats_sorted":["cs.AI","cs.MA"],"title_canon_sha256":"fc5600aec32d7565a760f8311e55f6712cb479366969b2fb3ba8c5d7a453736c","abstract_canon_sha256":"f1210784f75fa632740eb133c47a4642fae787aa2623e4f4d1a3d8b3af46de87"},"schema_version":"1.0"},"canonical_sha256":"f5453e8ff6046fc74f8fbed1e31bdc8c47f1716a5f3c0dcba4f07dae8ff81879","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T08:51:51.376179Z","signature_b64":"Vj0EeX9xmni4j76zbHCWJOssetcEVTUKHkJpChkIbhNFHyCOLeZ0RabSgGUMxyuOxqViGNTK/GW/ll8SbkJzBA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"f5453e8ff6046fc74f8fbed1e31bdc8c47f1716a5f3c0dcba4f07dae8ff81879","last_reissued_at":"2026-07-05T08:51:51.375676Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T08:51:51.375676Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2405.00823","source_version":2,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T08:51:51Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"E01mHVpb/qGf849ew+2xWn1Zn69LOjNlvcYsZs57ciGbOgpw3oy/YuBrXJr+YVm6eTia9LzkRjSglg1ATrLAAQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-08T06:24:19.892034Z"},"content_sha256":"5e2f0479c7e1c4069baabf7ea2d17ea45a30664091e5edab22db4a583552aa2a","schema_version":"1.0","event_id":"sha256:5e2f0479c7e1c4069baabf7ea2d17ea45a30664091e5edab22db4a583552aa2a"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2024:6VCT5D7WARX4OT4PX3I6GG64RR","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"WorkBench: a Benchmark Dataset for Agents in a Realistic Workplace Setting","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.MA"],"primary_cat":"cs.CL","authors_text":"Bertie Vidgen, Olly Styles, Patricio Cerda-Mardini, Sam Miller, Tanaya Guha, Victor Sanchez","submitted_at":"2024-05-01T19:07:03Z","abstract_excerpt":"We introduce WorkBench: a benchmark dataset for evaluating agents' ability to execute tasks in a workplace setting. WorkBench contains a sandbox environment with five databases, 26 tools, and 690 tasks. These tasks represent common business activities, such as sending emails and scheduling meetings. The tasks in WorkBench are challenging as they require planning, tool selection, and often multiple actions. If a task has been successfully executed, one (or more) of the database values may change. The correct outcome for each task is unique and unambiguous, which allows for robust, automated eva"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2405.00823","kind":"arxiv","version":2},"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/2405.00823/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":"2026-07-05T08:51:51Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"VhIPUXnJN005tPzIYmxVYU+ILdEGKqWhwpUI5S+gC1PEiggeBoqoDmdXwfGuo5kKo+ha3jEFcOYkSG3VR5B7Cw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-08T06:24:19.892925Z"},"content_sha256":"6f0712cd85394c0daf0f616ef3fc47ef6e181ea4bf796a3df5dfc9d415e22320","schema_version":"1.0","event_id":"sha256:6f0712cd85394c0daf0f616ef3fc47ef6e181ea4bf796a3df5dfc9d415e22320"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/6VCT5D7WARX4OT4PX3I6GG64RR/bundle.json","state_url":"https://pith.science/pith/6VCT5D7WARX4OT4PX3I6GG64RR/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/6VCT5D7WARX4OT4PX3I6GG64RR/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":"2026-08-08T06:24:19Z","links":{"resolver":"https://pith.science/pith/6VCT5D7WARX4OT4PX3I6GG64RR","bundle":"https://pith.science/pith/6VCT5D7WARX4OT4PX3I6GG64RR/bundle.json","state":"https://pith.science/pith/6VCT5D7WARX4OT4PX3I6GG64RR/state.json","well_known_bundle":"https://pith.science/.well-known/pith/6VCT5D7WARX4OT4PX3I6GG64RR/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:6VCT5D7WARX4OT4PX3I6GG64RR","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":"f1210784f75fa632740eb133c47a4642fae787aa2623e4f4d1a3d8b3af46de87","cross_cats_sorted":["cs.AI","cs.MA"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-05-01T19:07:03Z","title_canon_sha256":"fc5600aec32d7565a760f8311e55f6712cb479366969b2fb3ba8c5d7a453736c"},"schema_version":"1.0","source":{"id":"2405.00823","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2405.00823","created_at":"2026-07-05T08:51:51Z"},{"alias_kind":"arxiv_version","alias_value":"2405.00823v2","created_at":"2026-07-05T08:51:51Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2405.00823","created_at":"2026-07-05T08:51:51Z"},{"alias_kind":"pith_short_12","alias_value":"6VCT5D7WARX4","created_at":"2026-07-05T08:51:51Z"},{"alias_kind":"pith_short_16","alias_value":"6VCT5D7WARX4OT4P","created_at":"2026-07-05T08:51:51Z"},{"alias_kind":"pith_short_8","alias_value":"6VCT5D7W","created_at":"2026-07-05T08:51:51Z"}],"graph_snapshots":[{"event_id":"sha256:6f0712cd85394c0daf0f616ef3fc47ef6e181ea4bf796a3df5dfc9d415e22320","target":"graph","created_at":"2026-07-05T08:51:51Z","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/2405.00823/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"We introduce WorkBench: a benchmark dataset for evaluating agents' ability to execute tasks in a workplace setting. WorkBench contains a sandbox environment with five databases, 26 tools, and 690 tasks. These tasks represent common business activities, such as sending emails and scheduling meetings. The tasks in WorkBench are challenging as they require planning, tool selection, and often multiple actions. If a task has been successfully executed, one (or more) of the database values may change. The correct outcome for each task is unique and unambiguous, which allows for robust, automated eva","authors_text":"Bertie Vidgen, Olly Styles, Patricio Cerda-Mardini, Sam Miller, Tanaya Guha, Victor Sanchez","cross_cats":["cs.AI","cs.MA"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-05-01T19:07:03Z","title":"WorkBench: a Benchmark Dataset for Agents in a Realistic Workplace Setting"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2405.00823","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":"sha256:5e2f0479c7e1c4069baabf7ea2d17ea45a30664091e5edab22db4a583552aa2a","target":"record","created_at":"2026-07-05T08:51:51Z","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":"f1210784f75fa632740eb133c47a4642fae787aa2623e4f4d1a3d8b3af46de87","cross_cats_sorted":["cs.AI","cs.MA"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-05-01T19:07:03Z","title_canon_sha256":"fc5600aec32d7565a760f8311e55f6712cb479366969b2fb3ba8c5d7a453736c"},"schema_version":"1.0","source":{"id":"2405.00823","kind":"arxiv","version":2}},"canonical_sha256":"f5453e8ff6046fc74f8fbed1e31bdc8c47f1716a5f3c0dcba4f07dae8ff81879","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"f5453e8ff6046fc74f8fbed1e31bdc8c47f1716a5f3c0dcba4f07dae8ff81879","first_computed_at":"2026-07-05T08:51:51.375676Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T08:51:51.375676Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"Vj0EeX9xmni4j76zbHCWJOssetcEVTUKHkJpChkIbhNFHyCOLeZ0RabSgGUMxyuOxqViGNTK/GW/ll8SbkJzBA==","signature_status":"signed_v1","signed_at":"2026-07-05T08:51:51.376179Z","signed_message":"canonical_sha256_bytes"},"source_id":"2405.00823","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:5e2f0479c7e1c4069baabf7ea2d17ea45a30664091e5edab22db4a583552aa2a","sha256:6f0712cd85394c0daf0f616ef3fc47ef6e181ea4bf796a3df5dfc9d415e22320"],"state_sha256":"94345c37e2f96fe7554af53f08c996781642378e9426e5d48dea3ef989d98975"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"xjLqWSOdBAZJyAMiqF6gkSuFgykbyhbo0tJOeHd93kqwCD+LNgjaTVmM/jarETKy6DgCoR6vcCT0B+Y3AWmMAg==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-08T06:24:19.899305Z","bundle_sha256":"6dbe8ec2eb4b89ca5f4f1600ff4a20db516d65d84cca8b1f68f1ddb354a951a2"}}