{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2024:NRX2TYJYUMTP6YFWEVGNJDSUL5","short_pith_number":"pith:NRX2TYJY","canonical_record":{"source":{"id":"2411.02429","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2024-10-31T17:04:59Z","cross_cats_sorted":["cs.AI","cs.CE"],"title_canon_sha256":"8eba606f0f6e077864f3c2b1c99a14838d2ae1ac8d1194af47c1e0219917a719","abstract_canon_sha256":"8adce7bffd8ee23a75046a4dce0234ca48a904d8af5250d2091270f1197acd18"},"schema_version":"1.0"},"canonical_sha256":"6c6fa9e138a326ff60b6254cd48e545f4af43215412f93ede80696ebf546aa98","source":{"kind":"arxiv","id":"2411.02429","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2411.02429","created_at":"2026-07-05T09:31:11Z"},{"alias_kind":"arxiv_version","alias_value":"2411.02429v1","created_at":"2026-07-05T09:31:11Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2411.02429","created_at":"2026-07-05T09:31:11Z"},{"alias_kind":"pith_short_12","alias_value":"NRX2TYJYUMTP","created_at":"2026-07-05T09:31:11Z"},{"alias_kind":"pith_short_16","alias_value":"NRX2TYJYUMTP6YFW","created_at":"2026-07-05T09:31:11Z"},{"alias_kind":"pith_short_8","alias_value":"NRX2TYJY","created_at":"2026-07-05T09:31:11Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2024:NRX2TYJYUMTP6YFWEVGNJDSUL5","target":"record","payload":{"canonical_record":{"source":{"id":"2411.02429","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2024-10-31T17:04:59Z","cross_cats_sorted":["cs.AI","cs.CE"],"title_canon_sha256":"8eba606f0f6e077864f3c2b1c99a14838d2ae1ac8d1194af47c1e0219917a719","abstract_canon_sha256":"8adce7bffd8ee23a75046a4dce0234ca48a904d8af5250d2091270f1197acd18"},"schema_version":"1.0"},"canonical_sha256":"6c6fa9e138a326ff60b6254cd48e545f4af43215412f93ede80696ebf546aa98","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:31:11.682446Z","signature_b64":"RJ2qxfoAP+IDy+LSxVff9Zqo5Lsu/HDL0wQ90DAFvvqkUf2Goqo53HE/K3CpQ5v1r3+MhrlUqI6QhKIcWazABQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"6c6fa9e138a326ff60b6254cd48e545f4af43215412f93ede80696ebf546aa98","last_reissued_at":"2026-07-05T09:31:11.681980Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:31:11.681980Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2411.02429","source_version":1,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T09:31:11Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"3SolrLnppJ97Lh7COJyf9X5FVnPrbLmmvWwW1hJw4d1H9nug8IH5geFT1vACulsUsGHjDYdg0niBgNezBx1qCQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-10T18:06:07.095697Z"},"content_sha256":"566fa80db4750141004839dae77d9a81bc6282a0a19aea3de1a38189f3159e85","schema_version":"1.0","event_id":"sha256:566fa80db4750141004839dae77d9a81bc6282a0a19aea3de1a38189f3159e85"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2024:NRX2TYJYUMTP6YFWEVGNJDSUL5","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"IdeaBench: Benchmarking Large Language Models for Research Idea Generation","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":["cs.AI","cs.CE"],"primary_cat":"cs.CL","authors_text":"Aidong Zhang, Albert Huang, Amir Hassan Shariatmadari, Eric Xie, Guangzhi Xiong, Sikun Guo, Stefan Bekiranov","submitted_at":"2024-10-31T17:04:59Z","abstract_excerpt":"Large Language Models (LLMs) have transformed how people interact with artificial intelligence (AI) systems, achieving state-of-the-art results in various tasks, including scientific discovery and hypothesis generation. However, the lack of a comprehensive and systematic evaluation framework for generating research ideas using LLMs poses a significant obstacle to understanding and assessing their generative capabilities in scientific discovery. To address this gap, we propose IdeaBench, a benchmark system that includes a comprehensive dataset and an evaluation framework for standardizing the a"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2411.02429","kind":"arxiv","version":1},"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/2411.02429/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":"2026-07-05T09:31:11Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"LucPhbfw/dTeTLApRd+86txs7fB28mwVvwW4pRMYtc41lexonLWpEl5SNurOc3iF91JNZnQmHFkm34pKKKslAw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-10T18:06:07.096238Z"},"content_sha256":"12ebb66ea135c5ef885b194672c1bbdb58c09376cab7bd5da17ad20a79333810","schema_version":"1.0","event_id":"sha256:12ebb66ea135c5ef885b194672c1bbdb58c09376cab7bd5da17ad20a79333810"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/NRX2TYJYUMTP6YFWEVGNJDSUL5/bundle.json","state_url":"https://pith.science/pith/NRX2TYJYUMTP6YFWEVGNJDSUL5/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/NRX2TYJYUMTP6YFWEVGNJDSUL5/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":"2026-08-10T18:06:07Z","links":{"resolver":"https://pith.science/pith/NRX2TYJYUMTP6YFWEVGNJDSUL5","bundle":"https://pith.science/pith/NRX2TYJYUMTP6YFWEVGNJDSUL5/bundle.json","state":"https://pith.science/pith/NRX2TYJYUMTP6YFWEVGNJDSUL5/state.json","well_known_bundle":"https://pith.science/.well-known/pith/NRX2TYJYUMTP6YFWEVGNJDSUL5/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:NRX2TYJYUMTP6YFWEVGNJDSUL5","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":"8adce7bffd8ee23a75046a4dce0234ca48a904d8af5250d2091270f1197acd18","cross_cats_sorted":["cs.AI","cs.CE"],"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2024-10-31T17:04:59Z","title_canon_sha256":"8eba606f0f6e077864f3c2b1c99a14838d2ae1ac8d1194af47c1e0219917a719"},"schema_version":"1.0","source":{"id":"2411.02429","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2411.02429","created_at":"2026-07-05T09:31:11Z"},{"alias_kind":"arxiv_version","alias_value":"2411.02429v1","created_at":"2026-07-05T09:31:11Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2411.02429","created_at":"2026-07-05T09:31:11Z"},{"alias_kind":"pith_short_12","alias_value":"NRX2TYJYUMTP","created_at":"2026-07-05T09:31:11Z"},{"alias_kind":"pith_short_16","alias_value":"NRX2TYJYUMTP6YFW","created_at":"2026-07-05T09:31:11Z"},{"alias_kind":"pith_short_8","alias_value":"NRX2TYJY","created_at":"2026-07-05T09:31:11Z"}],"graph_snapshots":[{"event_id":"sha256:12ebb66ea135c5ef885b194672c1bbdb58c09376cab7bd5da17ad20a79333810","target":"graph","created_at":"2026-07-05T09:31:11Z","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/2411.02429/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Large Language Models (LLMs) have transformed how people interact with artificial intelligence (AI) systems, achieving state-of-the-art results in various tasks, including scientific discovery and hypothesis generation. However, the lack of a comprehensive and systematic evaluation framework for generating research ideas using LLMs poses a significant obstacle to understanding and assessing their generative capabilities in scientific discovery. To address this gap, we propose IdeaBench, a benchmark system that includes a comprehensive dataset and an evaluation framework for standardizing the a","authors_text":"Aidong Zhang, Albert Huang, Amir Hassan Shariatmadari, Eric Xie, Guangzhi Xiong, Sikun Guo, Stefan Bekiranov","cross_cats":["cs.AI","cs.CE"],"headline":"","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2024-10-31T17:04:59Z","title":"IdeaBench: Benchmarking Large Language Models for Research Idea Generation"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2411.02429","kind":"arxiv","version":1},"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":"sha256:566fa80db4750141004839dae77d9a81bc6282a0a19aea3de1a38189f3159e85","target":"record","created_at":"2026-07-05T09:31:11Z","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":"8adce7bffd8ee23a75046a4dce0234ca48a904d8af5250d2091270f1197acd18","cross_cats_sorted":["cs.AI","cs.CE"],"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2024-10-31T17:04:59Z","title_canon_sha256":"8eba606f0f6e077864f3c2b1c99a14838d2ae1ac8d1194af47c1e0219917a719"},"schema_version":"1.0","source":{"id":"2411.02429","kind":"arxiv","version":1}},"canonical_sha256":"6c6fa9e138a326ff60b6254cd48e545f4af43215412f93ede80696ebf546aa98","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"6c6fa9e138a326ff60b6254cd48e545f4af43215412f93ede80696ebf546aa98","first_computed_at":"2026-07-05T09:31:11.681980Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T09:31:11.681980Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"RJ2qxfoAP+IDy+LSxVff9Zqo5Lsu/HDL0wQ90DAFvvqkUf2Goqo53HE/K3CpQ5v1r3+MhrlUqI6QhKIcWazABQ==","signature_status":"signed_v1","signed_at":"2026-07-05T09:31:11.682446Z","signed_message":"canonical_sha256_bytes"},"source_id":"2411.02429","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:566fa80db4750141004839dae77d9a81bc6282a0a19aea3de1a38189f3159e85","sha256:12ebb66ea135c5ef885b194672c1bbdb58c09376cab7bd5da17ad20a79333810"],"state_sha256":"a0c5ca1d776fbf8a4a76e2eb70638228885382de38c1081a94047c85877326db"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"lAIFA8jcvqYfmua+cb/59PzoLCrR/vv3o+CyPseS62nkYDdbjrJmJ85MlPUwkMFxqGptIxAGzG6dDQ3llNjXAg==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-10T18:06:07.101813Z","bundle_sha256":"18870e7a9fa6648cb7530e89af582b6c1bda90dfb12518767f8c91ff3a55ee67"}}