{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:OQXPXYS2PJBDSLGDHZYKPYWIV6","short_pith_number":"pith:OQXPXYS2","schema_version":"1.0","canonical_sha256":"742efbe25a7a42392cc33e70a7e2c8af8133a762f80a8db5604c853bc4945de4","source":{"kind":"arxiv","id":"2312.12747","version":2},"attestation_state":"computed","paper":{"title":"ALMANACS: A Simulatability Benchmark for Language Model Explainability","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.CL","stat.ML"],"primary_cat":"cs.LG","authors_text":"Edmund Mills, Scott Emmons, Shiye Su, Stuart Russell","submitted_at":"2023-12-20T03:44:18Z","abstract_excerpt":"How do we measure the efficacy of language model explainability methods? While many explainability methods have been developed, they are typically evaluated on bespoke tasks, preventing an apples-to-apples comparison. To help fill this gap, we present ALMANACS, a language model explainability benchmark. ALMANACS scores explainability methods on simulatability, i.e., how well the explanations improve behavior prediction on new inputs. The ALMANACS scenarios span twelve safety-relevant topics such as ethical reasoning and advanced AI behaviors; they have idiosyncratic premises to invoke model-sp"},"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":"2312.12747","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2023-12-20T03:44:18Z","cross_cats_sorted":["cs.AI","cs.CL","stat.ML"],"title_canon_sha256":"bbd6d1e1f5e444a0ecc3a40c1db451ed850f74342ca7489e0354d16710feb8b0","abstract_canon_sha256":"56d49605aede8f0926d8f578f30a7dd02177e122117e5d71242453c03cda16be"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:08:13.398797Z","signature_b64":"2nV1YKF5zMVLeghIchbbqHx17CDnHyow0zZJBb9sidVYOlmv9BUtoUrCXfeWMrMqvxig9pD7ctvGp50oqJp9Dw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"742efbe25a7a42392cc33e70a7e2c8af8133a762f80a8db5604c853bc4945de4","last_reissued_at":"2026-07-05T10:08:13.398255Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:08:13.398255Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"ALMANACS: A Simulatability Benchmark for Language Model Explainability","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.CL","stat.ML"],"primary_cat":"cs.LG","authors_text":"Edmund Mills, Scott Emmons, Shiye Su, Stuart Russell","submitted_at":"2023-12-20T03:44:18Z","abstract_excerpt":"How do we measure the efficacy of language model explainability methods? While many explainability methods have been developed, they are typically evaluated on bespoke tasks, preventing an apples-to-apples comparison. To help fill this gap, we present ALMANACS, a language model explainability benchmark. ALMANACS scores explainability methods on simulatability, i.e., how well the explanations improve behavior prediction on new inputs. The ALMANACS scenarios span twelve safety-relevant topics such as ethical reasoning and advanced AI behaviors; they have idiosyncratic premises to invoke model-sp"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2312.12747","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/2312.12747/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":"2312.12747","created_at":"2026-07-05T10:08:13.398324+00:00"},{"alias_kind":"arxiv_version","alias_value":"2312.12747v2","created_at":"2026-07-05T10:08:13.398324+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2312.12747","created_at":"2026-07-05T10:08:13.398324+00:00"},{"alias_kind":"pith_short_12","alias_value":"OQXPXYS2PJBD","created_at":"2026-07-05T10:08:13.398324+00:00"},{"alias_kind":"pith_short_16","alias_value":"OQXPXYS2PJBDSLGD","created_at":"2026-07-05T10:08:13.398324+00:00"},{"alias_kind":"pith_short_8","alias_value":"OQXPXYS2","created_at":"2026-07-05T10:08:13.398324+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/OQXPXYS2PJBDSLGDHZYKPYWIV6","json":"https://pith.science/pith/OQXPXYS2PJBDSLGDHZYKPYWIV6.json","graph_json":"https://pith.science/api/pith-number/OQXPXYS2PJBDSLGDHZYKPYWIV6/graph.json","events_json":"https://pith.science/api/pith-number/OQXPXYS2PJBDSLGDHZYKPYWIV6/events.json","paper":"https://pith.science/paper/OQXPXYS2"},"agent_actions":{"view_html":"https://pith.science/pith/OQXPXYS2PJBDSLGDHZYKPYWIV6","download_json":"https://pith.science/pith/OQXPXYS2PJBDSLGDHZYKPYWIV6.json","view_paper":"https://pith.science/paper/OQXPXYS2","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2312.12747&json=true","fetch_graph":"https://pith.science/api/pith-number/OQXPXYS2PJBDSLGDHZYKPYWIV6/graph.json","fetch_events":"https://pith.science/api/pith-number/OQXPXYS2PJBDSLGDHZYKPYWIV6/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/OQXPXYS2PJBDSLGDHZYKPYWIV6/action/timestamp_anchor","attest_storage":"https://pith.science/pith/OQXPXYS2PJBDSLGDHZYKPYWIV6/action/storage_attestation","attest_author":"https://pith.science/pith/OQXPXYS2PJBDSLGDHZYKPYWIV6/action/author_attestation","sign_citation":"https://pith.science/pith/OQXPXYS2PJBDSLGDHZYKPYWIV6/action/citation_signature","submit_replication":"https://pith.science/pith/OQXPXYS2PJBDSLGDHZYKPYWIV6/action/replication_record"}},"created_at":"2026-07-05T10:08:13.398324+00:00","updated_at":"2026-07-05T10:08:13.398324+00:00"}