{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2025:IPT7SHSYGJ4OUHZ5MLVGWVY5GB","short_pith_number":"pith:IPT7SHSY","canonical_record":{"source":{"id":"2507.13300","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-07-17T17:09:22Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"70a936e91201f0f37e2f161eedb06b856db93b8165b99b3990a91fc8e42f4c3e","abstract_canon_sha256":"390a6d52281f617ea485aa3fcb0296a259dd74f8f4f33d878b8fa6151518961a"},"schema_version":"1.0"},"canonical_sha256":"43e7f91e583278ea1f3d62ea6b571d307a1cf049d8ef85e012e31a03d31b766c","source":{"kind":"arxiv","id":"2507.13300","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2507.13300","created_at":"2026-07-05T11:38:58Z"},{"alias_kind":"arxiv_version","alias_value":"2507.13300v1","created_at":"2026-07-05T11:38:58Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2507.13300","created_at":"2026-07-05T11:38:58Z"},{"alias_kind":"pith_short_12","alias_value":"IPT7SHSYGJ4O","created_at":"2026-07-05T11:38:58Z"},{"alias_kind":"pith_short_16","alias_value":"IPT7SHSYGJ4OUHZ5","created_at":"2026-07-05T11:38:58Z"},{"alias_kind":"pith_short_8","alias_value":"IPT7SHSY","created_at":"2026-07-05T11:38:58Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2025:IPT7SHSYGJ4OUHZ5MLVGWVY5GB","target":"record","payload":{"canonical_record":{"source":{"id":"2507.13300","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-07-17T17:09:22Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"70a936e91201f0f37e2f161eedb06b856db93b8165b99b3990a91fc8e42f4c3e","abstract_canon_sha256":"390a6d52281f617ea485aa3fcb0296a259dd74f8f4f33d878b8fa6151518961a"},"schema_version":"1.0"},"canonical_sha256":"43e7f91e583278ea1f3d62ea6b571d307a1cf049d8ef85e012e31a03d31b766c","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:38:58.722523Z","signature_b64":"lQhfUHagKywmW2fsN5NPRIr1OqSo0OP8PtTl7hifZwvGba7nhGbDgg7liBOaANCykdqluRTCuDcLnLhm6eS8AQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"43e7f91e583278ea1f3d62ea6b571d307a1cf049d8ef85e012e31a03d31b766c","last_reissued_at":"2026-07-05T11:38:58.722052Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:38:58.722052Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2507.13300","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":"2026-07-05T11:38:58Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"p6nUD+/IvPIv/3wcGWvWPDr8q4nzmLXT4h0UC6giBsrePMHCdAadJb03x9MeL3MY28LvxfQ5fKnMsGcqTuJaDw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-07T05:07:18.859685Z"},"content_sha256":"e4cd53dedeb4e3ec4c977bad229b66a652a47067cb8d8d6ed1a1afb9e928e98f","schema_version":"1.0","event_id":"sha256:e4cd53dedeb4e3ec4c977bad229b66a652a47067cb8d8d6ed1a1afb9e928e98f"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2025:IPT7SHSYGJ4OUHZ5MLVGWVY5GB","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"AbGen: Evaluating Large Language Models in Ablation Study Design and Evaluation for Scientific Research","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Arman Cohan, Chengye Wang, Lovekesh Vig, Manasi Patwardhan, Weiyuan Chen, Yilun Zhao, Yixin Liu, Zhijian Xu","submitted_at":"2025-07-17T17:09:22Z","abstract_excerpt":"We introduce AbGen, the first benchmark designed to evaluate the capabilities of LLMs in designing ablation studies for scientific research. AbGen consists of 1,500 expert-annotated examples derived from 807 NLP papers. In this benchmark, LLMs are tasked with generating detailed ablation study designs for a specified module or process based on the given research context. Our evaluation of leading LLMs, such as DeepSeek-R1-0528 and o4-mini, highlights a significant performance gap between these models and human experts in terms of the importance, faithfulness, and soundness of the ablation stud"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2507.13300","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/2507.13300/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":"2026-07-05T11:38:58Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"wC47ckkReu9nZZ/IlTB4D5A1T7tynZXCGO6DPTOt1a3rdTpilQzyqcglu1ZEd6rShcnHu/Bu2KbAny1aeDxqDQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-07T05:07:18.860270Z"},"content_sha256":"4d83613f9d60abda321746906c818b6eaa3f19476413206ec6acc6767c8969a5","schema_version":"1.0","event_id":"sha256:4d83613f9d60abda321746906c818b6eaa3f19476413206ec6acc6767c8969a5"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/IPT7SHSYGJ4OUHZ5MLVGWVY5GB/bundle.json","state_url":"https://pith.science/pith/IPT7SHSYGJ4OUHZ5MLVGWVY5GB/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/IPT7SHSYGJ4OUHZ5MLVGWVY5GB/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":"2026-08-07T05:07:18Z","links":{"resolver":"https://pith.science/pith/IPT7SHSYGJ4OUHZ5MLVGWVY5GB","bundle":"https://pith.science/pith/IPT7SHSYGJ4OUHZ5MLVGWVY5GB/bundle.json","state":"https://pith.science/pith/IPT7SHSYGJ4OUHZ5MLVGWVY5GB/state.json","well_known_bundle":"https://pith.science/.well-known/pith/IPT7SHSYGJ4OUHZ5MLVGWVY5GB/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:IPT7SHSYGJ4OUHZ5MLVGWVY5GB","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":"390a6d52281f617ea485aa3fcb0296a259dd74f8f4f33d878b8fa6151518961a","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-07-17T17:09:22Z","title_canon_sha256":"70a936e91201f0f37e2f161eedb06b856db93b8165b99b3990a91fc8e42f4c3e"},"schema_version":"1.0","source":{"id":"2507.13300","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2507.13300","created_at":"2026-07-05T11:38:58Z"},{"alias_kind":"arxiv_version","alias_value":"2507.13300v1","created_at":"2026-07-05T11:38:58Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2507.13300","created_at":"2026-07-05T11:38:58Z"},{"alias_kind":"pith_short_12","alias_value":"IPT7SHSYGJ4O","created_at":"2026-07-05T11:38:58Z"},{"alias_kind":"pith_short_16","alias_value":"IPT7SHSYGJ4OUHZ5","created_at":"2026-07-05T11:38:58Z"},{"alias_kind":"pith_short_8","alias_value":"IPT7SHSY","created_at":"2026-07-05T11:38:58Z"}],"graph_snapshots":[{"event_id":"sha256:4d83613f9d60abda321746906c818b6eaa3f19476413206ec6acc6767c8969a5","target":"graph","created_at":"2026-07-05T11:38:58Z","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/2507.13300/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"We introduce AbGen, the first benchmark designed to evaluate the capabilities of LLMs in designing ablation studies for scientific research. AbGen consists of 1,500 expert-annotated examples derived from 807 NLP papers. In this benchmark, LLMs are tasked with generating detailed ablation study designs for a specified module or process based on the given research context. Our evaluation of leading LLMs, such as DeepSeek-R1-0528 and o4-mini, highlights a significant performance gap between these models and human experts in terms of the importance, faithfulness, and soundness of the ablation stud","authors_text":"Arman Cohan, Chengye Wang, Lovekesh Vig, Manasi Patwardhan, Weiyuan Chen, Yilun Zhao, Yixin Liu, Zhijian Xu","cross_cats":["cs.AI"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-07-17T17:09:22Z","title":"AbGen: Evaluating Large Language Models in Ablation Study Design and Evaluation for Scientific Research"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2507.13300","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:e4cd53dedeb4e3ec4c977bad229b66a652a47067cb8d8d6ed1a1afb9e928e98f","target":"record","created_at":"2026-07-05T11:38:58Z","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":"390a6d52281f617ea485aa3fcb0296a259dd74f8f4f33d878b8fa6151518961a","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-07-17T17:09:22Z","title_canon_sha256":"70a936e91201f0f37e2f161eedb06b856db93b8165b99b3990a91fc8e42f4c3e"},"schema_version":"1.0","source":{"id":"2507.13300","kind":"arxiv","version":1}},"canonical_sha256":"43e7f91e583278ea1f3d62ea6b571d307a1cf049d8ef85e012e31a03d31b766c","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"43e7f91e583278ea1f3d62ea6b571d307a1cf049d8ef85e012e31a03d31b766c","first_computed_at":"2026-07-05T11:38:58.722052Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T11:38:58.722052Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"lQhfUHagKywmW2fsN5NPRIr1OqSo0OP8PtTl7hifZwvGba7nhGbDgg7liBOaANCykdqluRTCuDcLnLhm6eS8AQ==","signature_status":"signed_v1","signed_at":"2026-07-05T11:38:58.722523Z","signed_message":"canonical_sha256_bytes"},"source_id":"2507.13300","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:e4cd53dedeb4e3ec4c977bad229b66a652a47067cb8d8d6ed1a1afb9e928e98f","sha256:4d83613f9d60abda321746906c818b6eaa3f19476413206ec6acc6767c8969a5"],"state_sha256":"f1febc92bdaa45c1fe0a89965f882f09cb9362a944bb013fdcda1e381bf6fd72"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"BZqGtRTElj7bxynFFrr5ACUeEOzwcHbznIAF988m+s7L78VoVtGY8Fu1u9RkdupBfz0TSqmZ76umIFjVPkxlCw==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-07T05:07:18.864700Z","bundle_sha256":"7c98db89bb6dc7fed992325b86b7647768dcbaa5348df8c82935bee453c3bb30"}}