{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:C6MKQSQRSXYAEHMZNVUFF6NPGT","short_pith_number":"pith:C6MKQSQR","schema_version":"1.0","canonical_sha256":"1798a84a1195f0021d996d6852f9af34d4a4378c9014507c46b6f17b93e3c9df","source":{"kind":"arxiv","id":"2507.08038","version":3},"attestation_state":"computed","paper":{"title":"AblationBench: Evaluating Automated Planning of Ablations in Empirical AI Research","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Gal Chechik, Talor Abramovich","submitted_at":"2025-07-09T12:07:38Z","abstract_excerpt":"Language model agents are increasingly used to automate scientific research, yet evaluating their scientific contributions remains a challenge. A key mechanism to obtain such insights is through ablation experiments. To this end, we introduce AblationBench, a benchmark suite for evaluating agents on ablation planning tasks in empirical AI research. It includes two tasks: AuthorAblation, which helps authors propose ablation experiments based on a method section and contains 83 instances, and ReviewerAblation, which helps reviewers find missing ablations in a full paper and contains 350 instance"},"verification_status":{"content_addressed":true,"pith_receipt":true,"author_attested":false,"weak_author_claims":0,"strong_author_claims":0,"externally_anchored":false,"storage_verified":false,"citation_signatures":0,"replication_records":0,"graph_snapshot":true,"references_resolved":false,"formal_links_present":false},"canonical_record":{"source":{"id":"2507.08038","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-07-09T12:07:38Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"92750522bf2d371e44dfaf57b552f17593f5b1196ba366efa6559ebf6ddc8374","abstract_canon_sha256":"aa269110803a3558d96e143fb12fd7b99f2c66f150fe940305428457c52a48a9"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-02T02:04:47.421458Z","signature_b64":"oc6yf4DK2daVvUpZfJ3onPcU52IrnrQFkM3o36lgNbNR7ZxgxGxJLK0fe5z8ncimV1o/4OWGt6zV9uywCo8HBw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"1798a84a1195f0021d996d6852f9af34d4a4378c9014507c46b6f17b93e3c9df","last_reissued_at":"2026-06-02T02:04:47.420923Z","signature_status":"signed_v1","first_computed_at":"2026-06-02T02:04:47.420923Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"AblationBench: Evaluating Automated Planning of Ablations in Empirical AI Research","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Gal Chechik, Talor Abramovich","submitted_at":"2025-07-09T12:07:38Z","abstract_excerpt":"Language model agents are increasingly used to automate scientific research, yet evaluating their scientific contributions remains a challenge. A key mechanism to obtain such insights is through ablation experiments. To this end, we introduce AblationBench, a benchmark suite for evaluating agents on ablation planning tasks in empirical AI research. It includes two tasks: AuthorAblation, which helps authors propose ablation experiments based on a method section and contains 83 instances, and ReviewerAblation, which helps reviewers find missing ablations in a full paper and contains 350 instance"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2507.08038","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/2507.08038/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"},"aliases":[{"alias_kind":"arxiv","alias_value":"2507.08038","created_at":"2026-06-02T02:04:47.420989+00:00"},{"alias_kind":"arxiv_version","alias_value":"2507.08038v3","created_at":"2026-06-02T02:04:47.420989+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2507.08038","created_at":"2026-06-02T02:04:47.420989+00:00"},{"alias_kind":"pith_short_12","alias_value":"C6MKQSQRSXYA","created_at":"2026-06-02T02:04:47.420989+00:00"},{"alias_kind":"pith_short_16","alias_value":"C6MKQSQRSXYAEHMZ","created_at":"2026-06-02T02:04:47.420989+00:00"},{"alias_kind":"pith_short_8","alias_value":"C6MKQSQR","created_at":"2026-06-02T02:04:47.420989+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":2,"internal_anchor_count":2,"sample":[{"citing_arxiv_id":"2604.19606","citing_title":"AblateCell: A Reproduce-then-Ablate Agent for Virtual Cell Repositories","ref_index":31,"is_internal_anchor":true},{"citing_arxiv_id":"2604.20927","citing_title":"Hidden Secrets in the arXiv: Discovering, Analyzing, and Preventing Unintentional Information Disclosure in Source Files of Scientific Preprints","ref_index":112,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/C6MKQSQRSXYAEHMZNVUFF6NPGT","json":"https://pith.science/pith/C6MKQSQRSXYAEHMZNVUFF6NPGT.json","graph_json":"https://pith.science/api/pith-number/C6MKQSQRSXYAEHMZNVUFF6NPGT/graph.json","events_json":"https://pith.science/api/pith-number/C6MKQSQRSXYAEHMZNVUFF6NPGT/events.json","paper":"https://pith.science/paper/C6MKQSQR"},"agent_actions":{"view_html":"https://pith.science/pith/C6MKQSQRSXYAEHMZNVUFF6NPGT","download_json":"https://pith.science/pith/C6MKQSQRSXYAEHMZNVUFF6NPGT.json","view_paper":"https://pith.science/paper/C6MKQSQR","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2507.08038&json=true","fetch_graph":"https://pith.science/api/pith-number/C6MKQSQRSXYAEHMZNVUFF6NPGT/graph.json","fetch_events":"https://pith.science/api/pith-number/C6MKQSQRSXYAEHMZNVUFF6NPGT/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/C6MKQSQRSXYAEHMZNVUFF6NPGT/action/timestamp_anchor","attest_storage":"https://pith.science/pith/C6MKQSQRSXYAEHMZNVUFF6NPGT/action/storage_attestation","attest_author":"https://pith.science/pith/C6MKQSQRSXYAEHMZNVUFF6NPGT/action/author_attestation","sign_citation":"https://pith.science/pith/C6MKQSQRSXYAEHMZNVUFF6NPGT/action/citation_signature","submit_replication":"https://pith.science/pith/C6MKQSQRSXYAEHMZNVUFF6NPGT/action/replication_record"}},"created_at":"2026-06-02T02:04:47.420989+00:00","updated_at":"2026-06-02T02:04:47.420989+00:00"}