{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2022:F7USRYE464KHUQA6I4OSKFQPVA","short_pith_number":"pith:F7USRYE4","canonical_record":{"source":{"id":"2211.02069","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2022-11-03T18:01:12Z","cross_cats_sorted":["cs.AI","cs.LG"],"title_canon_sha256":"3b2c6b055b16f3d4100fbdf5d6ce2de8baf73ac9e97c4108a6ddef7c5525dbce","abstract_canon_sha256":"766f5850d725085cf1be10bc34258314a91897782f1b671a2db6c9f7850bee18"},"schema_version":"1.0"},"canonical_sha256":"2fe928e09cf7147a401e471d25160fa83960e5cf0b9c878a0168c5fede2dc6f6","source":{"kind":"arxiv","id":"2211.02069","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2211.02069","created_at":"2026-07-05T05:26:10Z"},{"alias_kind":"arxiv_version","alias_value":"2211.02069v2","created_at":"2026-07-05T05:26:10Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2211.02069","created_at":"2026-07-05T05:26:10Z"},{"alias_kind":"pith_short_12","alias_value":"F7USRYE464KH","created_at":"2026-07-05T05:26:10Z"},{"alias_kind":"pith_short_16","alias_value":"F7USRYE464KHUQA6","created_at":"2026-07-05T05:26:10Z"},{"alias_kind":"pith_short_8","alias_value":"F7USRYE4","created_at":"2026-07-05T05:26:10Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2022:F7USRYE464KHUQA6I4OSKFQPVA","target":"record","payload":{"canonical_record":{"source":{"id":"2211.02069","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2022-11-03T18:01:12Z","cross_cats_sorted":["cs.AI","cs.LG"],"title_canon_sha256":"3b2c6b055b16f3d4100fbdf5d6ce2de8baf73ac9e97c4108a6ddef7c5525dbce","abstract_canon_sha256":"766f5850d725085cf1be10bc34258314a91897782f1b671a2db6c9f7850bee18"},"schema_version":"1.0"},"canonical_sha256":"2fe928e09cf7147a401e471d25160fa83960e5cf0b9c878a0168c5fede2dc6f6","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T05:26:10.477471Z","signature_b64":"Alyuubg+uMi9kZsCz6HpQyb63xQFtRy2uXzTUq6EhMLZlVHMuMDQD/QTLTLwXC6FUr+zMsSjNVMzeleZIiZFBA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"2fe928e09cf7147a401e471d25160fa83960e5cf0b9c878a0168c5fede2dc6f6","last_reissued_at":"2026-07-05T05:26:10.477057Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T05:26:10.477057Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2211.02069","source_version":2,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T05:26:10Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"o2qe7KgKfHS/ka7EI/mnD/n5yJd51Bhtt3T4xVo1OgAn6RlVKHP12se/qKF4RlogIYL7B7WWiZkMdWWJgCMQDg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-15T21:12:29.424009Z"},"content_sha256":"e068d047d4ba7b39a147604c7973d0c9fa22f60c463f33e0c607c7c653982343","schema_version":"1.0","event_id":"sha256:e068d047d4ba7b39a147604c7973d0c9fa22f60c463f33e0c607c7c653982343"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2022:F7USRYE464KHUQA6I4OSKFQPVA","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"LMentry: A Language Model Benchmark of Elementary Language Tasks","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.LG"],"primary_cat":"cs.CL","authors_text":"Avia Efrat, Omer Levy, Or Honovich","submitted_at":"2022-11-03T18:01:12Z","abstract_excerpt":"As the performance of large language models rapidly improves, benchmarks are getting larger and more complex as well. We present LMentry, a benchmark that avoids this \"arms race\" by focusing on a compact set of tasks that are trivial to humans, e.g. writing a sentence containing a specific word, identifying which words in a list belong to a specific category, or choosing which of two words is longer. LMentry is specifically designed to provide quick and interpretable insights into the capabilities and robustness of large language models. Our experiments reveal a wide variety of failure cases t"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2211.02069","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/2211.02069/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":"2026-07-05T05:26:10Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"nL2Gz8GXKZlbsonZVOUK6QqQFfgLLJBWqW4VKWdS2yRJAGnaerAeyrKS2afIdIZOpZNwqcqwwWGI/8I75NWiAA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-15T21:12:29.424651Z"},"content_sha256":"fad4422082d3603fda10dd066470443012dd0e761968cbb3f606d39bca174377","schema_version":"1.0","event_id":"sha256:fad4422082d3603fda10dd066470443012dd0e761968cbb3f606d39bca174377"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/F7USRYE464KHUQA6I4OSKFQPVA/bundle.json","state_url":"https://pith.science/pith/F7USRYE464KHUQA6I4OSKFQPVA/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/F7USRYE464KHUQA6I4OSKFQPVA/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":"2026-08-15T21:12:29Z","links":{"resolver":"https://pith.science/pith/F7USRYE464KHUQA6I4OSKFQPVA","bundle":"https://pith.science/pith/F7USRYE464KHUQA6I4OSKFQPVA/bundle.json","state":"https://pith.science/pith/F7USRYE464KHUQA6I4OSKFQPVA/state.json","well_known_bundle":"https://pith.science/.well-known/pith/F7USRYE464KHUQA6I4OSKFQPVA/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2022:F7USRYE464KHUQA6I4OSKFQPVA","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":"766f5850d725085cf1be10bc34258314a91897782f1b671a2db6c9f7850bee18","cross_cats_sorted":["cs.AI","cs.LG"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2022-11-03T18:01:12Z","title_canon_sha256":"3b2c6b055b16f3d4100fbdf5d6ce2de8baf73ac9e97c4108a6ddef7c5525dbce"},"schema_version":"1.0","source":{"id":"2211.02069","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2211.02069","created_at":"2026-07-05T05:26:10Z"},{"alias_kind":"arxiv_version","alias_value":"2211.02069v2","created_at":"2026-07-05T05:26:10Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2211.02069","created_at":"2026-07-05T05:26:10Z"},{"alias_kind":"pith_short_12","alias_value":"F7USRYE464KH","created_at":"2026-07-05T05:26:10Z"},{"alias_kind":"pith_short_16","alias_value":"F7USRYE464KHUQA6","created_at":"2026-07-05T05:26:10Z"},{"alias_kind":"pith_short_8","alias_value":"F7USRYE4","created_at":"2026-07-05T05:26:10Z"}],"graph_snapshots":[{"event_id":"sha256:fad4422082d3603fda10dd066470443012dd0e761968cbb3f606d39bca174377","target":"graph","created_at":"2026-07-05T05:26:10Z","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/2211.02069/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"As the performance of large language models rapidly improves, benchmarks are getting larger and more complex as well. We present LMentry, a benchmark that avoids this \"arms race\" by focusing on a compact set of tasks that are trivial to humans, e.g. writing a sentence containing a specific word, identifying which words in a list belong to a specific category, or choosing which of two words is longer. LMentry is specifically designed to provide quick and interpretable insights into the capabilities and robustness of large language models. Our experiments reveal a wide variety of failure cases t","authors_text":"Avia Efrat, Omer Levy, Or Honovich","cross_cats":["cs.AI","cs.LG"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2022-11-03T18:01:12Z","title":"LMentry: A Language Model Benchmark of Elementary Language Tasks"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2211.02069","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:e068d047d4ba7b39a147604c7973d0c9fa22f60c463f33e0c607c7c653982343","target":"record","created_at":"2026-07-05T05:26:10Z","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":"766f5850d725085cf1be10bc34258314a91897782f1b671a2db6c9f7850bee18","cross_cats_sorted":["cs.AI","cs.LG"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2022-11-03T18:01:12Z","title_canon_sha256":"3b2c6b055b16f3d4100fbdf5d6ce2de8baf73ac9e97c4108a6ddef7c5525dbce"},"schema_version":"1.0","source":{"id":"2211.02069","kind":"arxiv","version":2}},"canonical_sha256":"2fe928e09cf7147a401e471d25160fa83960e5cf0b9c878a0168c5fede2dc6f6","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"2fe928e09cf7147a401e471d25160fa83960e5cf0b9c878a0168c5fede2dc6f6","first_computed_at":"2026-07-05T05:26:10.477057Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T05:26:10.477057Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"Alyuubg+uMi9kZsCz6HpQyb63xQFtRy2uXzTUq6EhMLZlVHMuMDQD/QTLTLwXC6FUr+zMsSjNVMzeleZIiZFBA==","signature_status":"signed_v1","signed_at":"2026-07-05T05:26:10.477471Z","signed_message":"canonical_sha256_bytes"},"source_id":"2211.02069","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:e068d047d4ba7b39a147604c7973d0c9fa22f60c463f33e0c607c7c653982343","sha256:fad4422082d3603fda10dd066470443012dd0e761968cbb3f606d39bca174377"],"state_sha256":"be02591b554bf5304c52cb80a02a440a13b5df0158e6ae26322f59a5f2e236c9"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"P8iwP0dQpibLYk90uUaBoXQtBkVR2WxyfJQCkXt0krq21sf/cOCujjQOJ/Bz6zvKvhU7O+ehV3sk7YBmo8WpDA==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-15T21:12:29.430412Z","bundle_sha256":"e9366370d4b7313677793a03d2a18f9f66ad0458977af5b5e0c0a92380a9b2b6"}}