{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2024:YZV72VZW4RF3CYA32WT76S4HT7","short_pith_number":"pith:YZV72VZW","canonical_record":{"source":{"id":"2410.05080","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2024-10-07T14:33:50Z","cross_cats_sorted":["cs.AI","cs.LG"],"title_canon_sha256":"b1a53ca8c518fb7a6f242f17bf78f3d1a8b2ca172779a99d7a95bf6eff49d05c","abstract_canon_sha256":"1db686719223d2d476d019c136f90496173651dbd87de83cd9eca894666cc2d1"},"schema_version":"1.0"},"canonical_sha256":"c66bfd5736e44bb1601bd5a7ff4b879ff39fd9be8faec7bad631924589dfecf0","source":{"kind":"arxiv","id":"2410.05080","version":3},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2410.05080","created_at":"2026-07-05T10:41:49Z"},{"alias_kind":"arxiv_version","alias_value":"2410.05080v3","created_at":"2026-07-05T10:41:49Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2410.05080","created_at":"2026-07-05T10:41:49Z"},{"alias_kind":"pith_short_12","alias_value":"YZV72VZW4RF3","created_at":"2026-07-05T10:41:49Z"},{"alias_kind":"pith_short_16","alias_value":"YZV72VZW4RF3CYA3","created_at":"2026-07-05T10:41:49Z"},{"alias_kind":"pith_short_8","alias_value":"YZV72VZW","created_at":"2026-07-05T10:41:49Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2024:YZV72VZW4RF3CYA32WT76S4HT7","target":"record","payload":{"canonical_record":{"source":{"id":"2410.05080","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2024-10-07T14:33:50Z","cross_cats_sorted":["cs.AI","cs.LG"],"title_canon_sha256":"b1a53ca8c518fb7a6f242f17bf78f3d1a8b2ca172779a99d7a95bf6eff49d05c","abstract_canon_sha256":"1db686719223d2d476d019c136f90496173651dbd87de83cd9eca894666cc2d1"},"schema_version":"1.0"},"canonical_sha256":"c66bfd5736e44bb1601bd5a7ff4b879ff39fd9be8faec7bad631924589dfecf0","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:41:49.046399Z","signature_b64":"8B75A7SB+YM63pukt2leG3OQwfQ9sKk5rrqQHXx9TPvGRlIL6D0eVNi8gnQdVw/m4O/68yvmbAJdU3okdDaBAQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"c66bfd5736e44bb1601bd5a7ff4b879ff39fd9be8faec7bad631924589dfecf0","last_reissued_at":"2026-07-05T10:41:49.045900Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:41:49.045900Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2410.05080","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":"2026-07-05T10:41:49Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"a/lj1R3nEC9hRLN7jHmEobnQG48Hu+blG8vs5uI3lR1VXQKFUYf/h2PeP9/PTCk2V9bUxnkiaSYy7q/oTXv/Bg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-01T07:54:29.837511Z"},"content_sha256":"2ae18cedab8b9313f7513a9c17567fa39e1dfbf59069b464eafce8eb084233c4","schema_version":"1.0","event_id":"sha256:2ae18cedab8b9313f7513a9c17567fa39e1dfbf59069b464eafce8eb084233c4"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2024:YZV72VZW4RF3CYA32WT76S4HT7","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"ScienceAgentBench: Toward Rigorous Assessment of Language Agents for Data-Driven Scientific Discovery","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.LG"],"primary_cat":"cs.CL","authors_text":"Benjamin Burns, Boshi Wang, Botao Yu, Chen Wei, Daniel Adu-Ampratwum, Frazier N. Baker, Huan Sun, Mingyi Xue, Qianheng Zhang, Shijie Chen, Song Gao, Vishal Dey, Xia Ning, Xuhui Huang, Yifei Li, Yu Su, Yuting Ning, Zeyi Liao, Ziru Chen, Zitong Lu","submitted_at":"2024-10-07T14:33:50Z","abstract_excerpt":"The advancements of large language models (LLMs) have piqued growing interest in developing LLM-based language agents to automate scientific discovery end-to-end, which has sparked both excitement and skepticism about their true capabilities. In this work, we call for rigorous assessment of agents on individual tasks in a scientific workflow before making bold claims on end-to-end automation. To this end, we present ScienceAgentBench, a new benchmark for evaluating language agents for data-driven scientific discovery. To ensure the scientific authenticity and real-world relevance of our benchm"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2410.05080","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/2410.05080/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":"2026-07-05T10:41:49Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"HMaqgZ6/FrkOo2+O+8lOZ8I16qo2moHa3XNXzkJjSW6XV3A9i8VHK2MlCJ83/8W/4qbIMpEAQ4a3Nr+/i0RMAw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-01T07:54:29.838032Z"},"content_sha256":"d75248613de1930cabec8180632abf3d99949035088745e3632857309b64bba0","schema_version":"1.0","event_id":"sha256:d75248613de1930cabec8180632abf3d99949035088745e3632857309b64bba0"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/YZV72VZW4RF3CYA32WT76S4HT7/bundle.json","state_url":"https://pith.science/pith/YZV72VZW4RF3CYA32WT76S4HT7/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/YZV72VZW4RF3CYA32WT76S4HT7/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":"2026-08-01T07:54:29Z","links":{"resolver":"https://pith.science/pith/YZV72VZW4RF3CYA32WT76S4HT7","bundle":"https://pith.science/pith/YZV72VZW4RF3CYA32WT76S4HT7/bundle.json","state":"https://pith.science/pith/YZV72VZW4RF3CYA32WT76S4HT7/state.json","well_known_bundle":"https://pith.science/.well-known/pith/YZV72VZW4RF3CYA32WT76S4HT7/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:YZV72VZW4RF3CYA32WT76S4HT7","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":"1db686719223d2d476d019c136f90496173651dbd87de83cd9eca894666cc2d1","cross_cats_sorted":["cs.AI","cs.LG"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2024-10-07T14:33:50Z","title_canon_sha256":"b1a53ca8c518fb7a6f242f17bf78f3d1a8b2ca172779a99d7a95bf6eff49d05c"},"schema_version":"1.0","source":{"id":"2410.05080","kind":"arxiv","version":3}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2410.05080","created_at":"2026-07-05T10:41:49Z"},{"alias_kind":"arxiv_version","alias_value":"2410.05080v3","created_at":"2026-07-05T10:41:49Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2410.05080","created_at":"2026-07-05T10:41:49Z"},{"alias_kind":"pith_short_12","alias_value":"YZV72VZW4RF3","created_at":"2026-07-05T10:41:49Z"},{"alias_kind":"pith_short_16","alias_value":"YZV72VZW4RF3CYA3","created_at":"2026-07-05T10:41:49Z"},{"alias_kind":"pith_short_8","alias_value":"YZV72VZW","created_at":"2026-07-05T10:41:49Z"}],"graph_snapshots":[{"event_id":"sha256:d75248613de1930cabec8180632abf3d99949035088745e3632857309b64bba0","target":"graph","created_at":"2026-07-05T10:41:49Z","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/2410.05080/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"The advancements of large language models (LLMs) have piqued growing interest in developing LLM-based language agents to automate scientific discovery end-to-end, which has sparked both excitement and skepticism about their true capabilities. In this work, we call for rigorous assessment of agents on individual tasks in a scientific workflow before making bold claims on end-to-end automation. To this end, we present ScienceAgentBench, a new benchmark for evaluating language agents for data-driven scientific discovery. To ensure the scientific authenticity and real-world relevance of our benchm","authors_text":"Benjamin Burns, Boshi Wang, Botao Yu, Chen Wei, Daniel Adu-Ampratwum, Frazier N. Baker, Huan Sun, Mingyi Xue, Qianheng Zhang, Shijie Chen, Song Gao, Vishal Dey, Xia Ning, Xuhui Huang, Yifei Li, Yu Su, Yuting Ning, Zeyi Liao, Ziru Chen, Zitong Lu","cross_cats":["cs.AI","cs.LG"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2024-10-07T14:33:50Z","title":"ScienceAgentBench: Toward Rigorous Assessment of Language Agents for Data-Driven Scientific Discovery"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2410.05080","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":"sha256:2ae18cedab8b9313f7513a9c17567fa39e1dfbf59069b464eafce8eb084233c4","target":"record","created_at":"2026-07-05T10:41:49Z","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":"1db686719223d2d476d019c136f90496173651dbd87de83cd9eca894666cc2d1","cross_cats_sorted":["cs.AI","cs.LG"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2024-10-07T14:33:50Z","title_canon_sha256":"b1a53ca8c518fb7a6f242f17bf78f3d1a8b2ca172779a99d7a95bf6eff49d05c"},"schema_version":"1.0","source":{"id":"2410.05080","kind":"arxiv","version":3}},"canonical_sha256":"c66bfd5736e44bb1601bd5a7ff4b879ff39fd9be8faec7bad631924589dfecf0","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"c66bfd5736e44bb1601bd5a7ff4b879ff39fd9be8faec7bad631924589dfecf0","first_computed_at":"2026-07-05T10:41:49.045900Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T10:41:49.045900Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"8B75A7SB+YM63pukt2leG3OQwfQ9sKk5rrqQHXx9TPvGRlIL6D0eVNi8gnQdVw/m4O/68yvmbAJdU3okdDaBAQ==","signature_status":"signed_v1","signed_at":"2026-07-05T10:41:49.046399Z","signed_message":"canonical_sha256_bytes"},"source_id":"2410.05080","source_kind":"arxiv","source_version":3}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:2ae18cedab8b9313f7513a9c17567fa39e1dfbf59069b464eafce8eb084233c4","sha256:d75248613de1930cabec8180632abf3d99949035088745e3632857309b64bba0"],"state_sha256":"6eb7ede68c2f86a345d97badf75a7e3c22a0ae26379558432f3c2a2b9f89da35"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"4RVZ48uVr1E/QqgXsQ9YW3D0Z6NdquuAryFEKLk87zTBQidxuCWjjbbaCtw1Xvlrgw2nW2DRdN73yeEMYdIRBg==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-01T07:54:29.841327Z","bundle_sha256":"1cebbdf8437567bb8424fd55582a461db2eabe46e9633c2af2292d8f02140946"}}