{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2024:UVDRQDL7FWQ6QBXTPZQ36PZZNO","short_pith_number":"pith:UVDRQDL7","canonical_record":{"source":{"id":"2409.15334","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-09-08T11:30:03Z","cross_cats_sorted":[],"title_canon_sha256":"34af3a610e59a331766e712f8e6221dc4c9789750c289c7e4f41d856619678ee","abstract_canon_sha256":"e7982a02f6bf9bc0901652a5abe7198b017784699b6e9e06c6910aa534e54a6b"},"schema_version":"1.0"},"canonical_sha256":"a547180d7f2da1e806f37e61bf3f396b9d3c1073b298b2c986e5be20545d55fd","source":{"kind":"arxiv","id":"2409.15334","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2409.15334","created_at":"2026-07-05T09:10:53Z"},{"alias_kind":"arxiv_version","alias_value":"2409.15334v1","created_at":"2026-07-05T09:10:53Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2409.15334","created_at":"2026-07-05T09:10:53Z"},{"alias_kind":"pith_short_12","alias_value":"UVDRQDL7FWQ6","created_at":"2026-07-05T09:10:53Z"},{"alias_kind":"pith_short_16","alias_value":"UVDRQDL7FWQ6QBXT","created_at":"2026-07-05T09:10:53Z"},{"alias_kind":"pith_short_8","alias_value":"UVDRQDL7","created_at":"2026-07-05T09:10:53Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2024:UVDRQDL7FWQ6QBXTPZQ36PZZNO","target":"record","payload":{"canonical_record":{"source":{"id":"2409.15334","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-09-08T11:30:03Z","cross_cats_sorted":[],"title_canon_sha256":"34af3a610e59a331766e712f8e6221dc4c9789750c289c7e4f41d856619678ee","abstract_canon_sha256":"e7982a02f6bf9bc0901652a5abe7198b017784699b6e9e06c6910aa534e54a6b"},"schema_version":"1.0"},"canonical_sha256":"a547180d7f2da1e806f37e61bf3f396b9d3c1073b298b2c986e5be20545d55fd","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:10:53.249776Z","signature_b64":"TIMhpoJcDrp7pl5VU79RI08V4e7+UQhzKlF1Hyx7MQQQti25uq2ipULlkNt5DgqJs9T2jIgtN+fukqiDNOicBA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"a547180d7f2da1e806f37e61bf3f396b9d3c1073b298b2c986e5be20545d55fd","last_reissued_at":"2026-07-05T09:10:53.249297Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:10:53.249297Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2409.15334","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":"2026-07-05T09:10:53Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"VKKzATU8ZcqpCwU8Xx/IqKy275b8PfQQZGX48F0leFXrrUxDFEagtSY8na4pX89rCYRBImz6PXEfVVzMhjJ1Cw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-04T13:45:14.997778Z"},"content_sha256":"e1970a11708c8beab02246aae0c098399af13cb43a50f214c53f8e340e8f814c","schema_version":"1.0","event_id":"sha256:e1970a11708c8beab02246aae0c098399af13cb43a50f214c53f8e340e8f814c"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2024:UVDRQDL7FWQ6QBXTPZQ36PZZNO","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Evaluating Large Language Models with Tests of Spanish as a Foreign Language: Pass or Fail?","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Elena Merino-G\\'omez, Javier Conde, Mar\\'ia Grandury, Marina Mayor-Rocher, Nina Melero, Pedro Reviriego","submitted_at":"2024-09-08T11:30:03Z","abstract_excerpt":"Large Language Models (LLMs) have been profusely evaluated on their ability to answer questions on many topics and their performance on different natural language understanding tasks. Those tests are usually conducted in English, but most LLM users are not native English speakers. Therefore, it is of interest to analyze how LLMs understand other languages at different levels: from paragraphs to morphems. In this paper, we evaluate the performance of state-of-the-art LLMs in TELEIA, a recently released benchmark with similar questions to those of Spanish exams for foreign students, covering top"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2409.15334","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/2409.15334/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":"2026-07-05T09:10:53Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"tXkT7fSJAuw86/dU9DUA5DxboZcJ1O3hWGHSliDeC5oAyNKWQF2zZyrfVrVLLeFiJ3rotwq/EDlOyQ1EfdXHBw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-04T13:45:14.998267Z"},"content_sha256":"156f907936e6a3214898ff6cb9bb26d2ecfd473c50beca5688088fba21160bbf","schema_version":"1.0","event_id":"sha256:156f907936e6a3214898ff6cb9bb26d2ecfd473c50beca5688088fba21160bbf"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/UVDRQDL7FWQ6QBXTPZQ36PZZNO/bundle.json","state_url":"https://pith.science/pith/UVDRQDL7FWQ6QBXTPZQ36PZZNO/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/UVDRQDL7FWQ6QBXTPZQ36PZZNO/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":"2026-08-04T13:45:15Z","links":{"resolver":"https://pith.science/pith/UVDRQDL7FWQ6QBXTPZQ36PZZNO","bundle":"https://pith.science/pith/UVDRQDL7FWQ6QBXTPZQ36PZZNO/bundle.json","state":"https://pith.science/pith/UVDRQDL7FWQ6QBXTPZQ36PZZNO/state.json","well_known_bundle":"https://pith.science/.well-known/pith/UVDRQDL7FWQ6QBXTPZQ36PZZNO/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:UVDRQDL7FWQ6QBXTPZQ36PZZNO","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":"e7982a02f6bf9bc0901652a5abe7198b017784699b6e9e06c6910aa534e54a6b","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-09-08T11:30:03Z","title_canon_sha256":"34af3a610e59a331766e712f8e6221dc4c9789750c289c7e4f41d856619678ee"},"schema_version":"1.0","source":{"id":"2409.15334","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2409.15334","created_at":"2026-07-05T09:10:53Z"},{"alias_kind":"arxiv_version","alias_value":"2409.15334v1","created_at":"2026-07-05T09:10:53Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2409.15334","created_at":"2026-07-05T09:10:53Z"},{"alias_kind":"pith_short_12","alias_value":"UVDRQDL7FWQ6","created_at":"2026-07-05T09:10:53Z"},{"alias_kind":"pith_short_16","alias_value":"UVDRQDL7FWQ6QBXT","created_at":"2026-07-05T09:10:53Z"},{"alias_kind":"pith_short_8","alias_value":"UVDRQDL7","created_at":"2026-07-05T09:10:53Z"}],"graph_snapshots":[{"event_id":"sha256:156f907936e6a3214898ff6cb9bb26d2ecfd473c50beca5688088fba21160bbf","target":"graph","created_at":"2026-07-05T09:10:53Z","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/2409.15334/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Large Language Models (LLMs) have been profusely evaluated on their ability to answer questions on many topics and their performance on different natural language understanding tasks. Those tests are usually conducted in English, but most LLM users are not native English speakers. Therefore, it is of interest to analyze how LLMs understand other languages at different levels: from paragraphs to morphems. In this paper, we evaluate the performance of state-of-the-art LLMs in TELEIA, a recently released benchmark with similar questions to those of Spanish exams for foreign students, covering top","authors_text":"Elena Merino-G\\'omez, Javier Conde, Mar\\'ia Grandury, Marina Mayor-Rocher, Nina Melero, Pedro Reviriego","cross_cats":[],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-09-08T11:30:03Z","title":"Evaluating Large Language Models with Tests of Spanish as a Foreign Language: Pass or Fail?"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2409.15334","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:e1970a11708c8beab02246aae0c098399af13cb43a50f214c53f8e340e8f814c","target":"record","created_at":"2026-07-05T09:10:53Z","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":"e7982a02f6bf9bc0901652a5abe7198b017784699b6e9e06c6910aa534e54a6b","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-09-08T11:30:03Z","title_canon_sha256":"34af3a610e59a331766e712f8e6221dc4c9789750c289c7e4f41d856619678ee"},"schema_version":"1.0","source":{"id":"2409.15334","kind":"arxiv","version":1}},"canonical_sha256":"a547180d7f2da1e806f37e61bf3f396b9d3c1073b298b2c986e5be20545d55fd","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"a547180d7f2da1e806f37e61bf3f396b9d3c1073b298b2c986e5be20545d55fd","first_computed_at":"2026-07-05T09:10:53.249297Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T09:10:53.249297Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"TIMhpoJcDrp7pl5VU79RI08V4e7+UQhzKlF1Hyx7MQQQti25uq2ipULlkNt5DgqJs9T2jIgtN+fukqiDNOicBA==","signature_status":"signed_v1","signed_at":"2026-07-05T09:10:53.249776Z","signed_message":"canonical_sha256_bytes"},"source_id":"2409.15334","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:e1970a11708c8beab02246aae0c098399af13cb43a50f214c53f8e340e8f814c","sha256:156f907936e6a3214898ff6cb9bb26d2ecfd473c50beca5688088fba21160bbf"],"state_sha256":"f787fb7bacd5d9b32c727991c87b9f26599ea24ba08b70d3e4189c1b57897d9f"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"Pgjk2vS03fMUW1E7rhTB0FVDoHv1clYvlRhpYDX8HwB064JzEHFjqs2ut0fUobfvrZagd9sjO01VhdkgZDJyDg==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-04T13:45:15.002199Z","bundle_sha256":"d07a22dd20d1111d96c5c20c034cc71b2385c0d7862e96c47f10ce413b875828"}}