{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2023:OQXPXYS2PJBDSLGDHZYKPYWIV6","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":"56d49605aede8f0926d8f578f30a7dd02177e122117e5d71242453c03cda16be","cross_cats_sorted":["cs.AI","cs.CL","stat.ML"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2023-12-20T03:44:18Z","title_canon_sha256":"bbd6d1e1f5e444a0ecc3a40c1db451ed850f74342ca7489e0354d16710feb8b0"},"schema_version":"1.0","source":{"id":"2312.12747","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2312.12747","created_at":"2026-07-05T10:08:13Z"},{"alias_kind":"arxiv_version","alias_value":"2312.12747v2","created_at":"2026-07-05T10:08:13Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2312.12747","created_at":"2026-07-05T10:08:13Z"},{"alias_kind":"pith_short_12","alias_value":"OQXPXYS2PJBD","created_at":"2026-07-05T10:08:13Z"},{"alias_kind":"pith_short_16","alias_value":"OQXPXYS2PJBDSLGD","created_at":"2026-07-05T10:08:13Z"},{"alias_kind":"pith_short_8","alias_value":"OQXPXYS2","created_at":"2026-07-05T10:08:13Z"}],"graph_snapshots":[{"event_id":"sha256:53a0518f59805db587c76bfd69891af1dd5660791e5d344c25f8c25bd16698a7","target":"graph","created_at":"2026-07-05T10:08:13Z","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/2312.12747/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"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","authors_text":"Edmund Mills, Scott Emmons, Shiye Su, Stuart Russell","cross_cats":["cs.AI","cs.CL","stat.ML"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2023-12-20T03:44:18Z","title":"ALMANACS: A Simulatability Benchmark for Language Model Explainability"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2312.12747","kind":"arxiv","version":2},"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":"sha256:239a3a847fba396541a83df80506aea363f18708fc3028329f8a44603b08d2e7","target":"record","created_at":"2026-07-05T10:08:13Z","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":"56d49605aede8f0926d8f578f30a7dd02177e122117e5d71242453c03cda16be","cross_cats_sorted":["cs.AI","cs.CL","stat.ML"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2023-12-20T03:44:18Z","title_canon_sha256":"bbd6d1e1f5e444a0ecc3a40c1db451ed850f74342ca7489e0354d16710feb8b0"},"schema_version":"1.0","source":{"id":"2312.12747","kind":"arxiv","version":2}},"canonical_sha256":"742efbe25a7a42392cc33e70a7e2c8af8133a762f80a8db5604c853bc4945de4","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"742efbe25a7a42392cc33e70a7e2c8af8133a762f80a8db5604c853bc4945de4","first_computed_at":"2026-07-05T10:08:13.398255Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T10:08:13.398255Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"2nV1YKF5zMVLeghIchbbqHx17CDnHyow0zZJBb9sidVYOlmv9BUtoUrCXfeWMrMqvxig9pD7ctvGp50oqJp9Dw==","signature_status":"signed_v1","signed_at":"2026-07-05T10:08:13.398797Z","signed_message":"canonical_sha256_bytes"},"source_id":"2312.12747","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:239a3a847fba396541a83df80506aea363f18708fc3028329f8a44603b08d2e7","sha256:53a0518f59805db587c76bfd69891af1dd5660791e5d344c25f8c25bd16698a7"],"state_sha256":"d566ef0917257a00c3fa17f32d05494e6da097f5c49a03057d75446a18a76ddb"}