{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2025:CCAPLEAOYNSJ7KQRVR6VCZ6AJH","short_pith_number":"pith:CCAPLEAO","canonical_record":{"source":{"id":"2509.07135","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-09-08T18:39:35Z","cross_cats_sorted":[],"title_canon_sha256":"db0abb7e10d7a9e6e277b22f0f208b71e8895d891280304cc78235675f33e297","abstract_canon_sha256":"1ba31cd11022a75d96f4c6de560718852bab84f5d195bed1a4d34a0332aded85"},"schema_version":"1.0"},"canonical_sha256":"1080f5900ec3649faa11ac7d5167c049f94b761828875ffd02aabf8ec6958aaa","source":{"kind":"arxiv","id":"2509.07135","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2509.07135","created_at":"2026-07-05T12:07:16Z"},{"alias_kind":"arxiv_version","alias_value":"2509.07135v1","created_at":"2026-07-05T12:07:16Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2509.07135","created_at":"2026-07-05T12:07:16Z"},{"alias_kind":"pith_short_12","alias_value":"CCAPLEAOYNSJ","created_at":"2026-07-05T12:07:16Z"},{"alias_kind":"pith_short_16","alias_value":"CCAPLEAOYNSJ7KQR","created_at":"2026-07-05T12:07:16Z"},{"alias_kind":"pith_short_8","alias_value":"CCAPLEAO","created_at":"2026-07-05T12:07:16Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2025:CCAPLEAOYNSJ7KQRVR6VCZ6AJH","target":"record","payload":{"canonical_record":{"source":{"id":"2509.07135","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-09-08T18:39:35Z","cross_cats_sorted":[],"title_canon_sha256":"db0abb7e10d7a9e6e277b22f0f208b71e8895d891280304cc78235675f33e297","abstract_canon_sha256":"1ba31cd11022a75d96f4c6de560718852bab84f5d195bed1a4d34a0332aded85"},"schema_version":"1.0"},"canonical_sha256":"1080f5900ec3649faa11ac7d5167c049f94b761828875ffd02aabf8ec6958aaa","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T12:07:16.147429Z","signature_b64":"PD14APgaEamnRuSkGpPBcaednC9m4vILF1eX8uvvt7hUmEnNcSOgM3ohRPAYnZGppa6t341o7vo4TzSbGrhSBw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"1080f5900ec3649faa11ac7d5167c049f94b761828875ffd02aabf8ec6958aaa","last_reissued_at":"2026-07-05T12:07:16.146914Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T12:07:16.146914Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2509.07135","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":"2026-07-05T12:07:16Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"iue75At9migHhb5UFDLRTDsiLB6Ia2Y4tsu0KZsBc4N6FWTRIWhXCh5AUP3FjjLnzUWLmhgAQXzjzXqKgbOUDA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-22T22:41:44.886552Z"},"content_sha256":"ef9a6ba4c61f5f12f5d545507524ecbe623271da58dbbf6c9b8cf2eeb8cc7cd3","schema_version":"1.0","event_id":"sha256:ef9a6ba4c61f5f12f5d545507524ecbe623271da58dbbf6c9b8cf2eeb8cc7cd3"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2025:CCAPLEAOYNSJ7KQRVR6VCZ6AJH","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"MedBench-IT: A Comprehensive Benchmark for Evaluating Large Language Models on Italian Medical Entrance Examinations","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Alessandro Angioi, Davide Sanna, Michelangelo Puliga, Roberto Marras, Ruggero Marino Lazzaroni","submitted_at":"2025-09-08T18:39:35Z","abstract_excerpt":"Large language models (LLMs) show increasing potential in education, yet benchmarks for non-English languages in specialized domains remain scarce. We introduce MedBench-IT, the first comprehensive benchmark for evaluating LLMs on Italian medical university entrance examinations. Sourced from Edizioni Simone, a leading preparatory materials publisher, MedBench-IT comprises 17,410 expert-written multiple-choice questions across six subjects (Biology, Chemistry, Logic, General Culture, Mathematics, Physics) and three difficulty levels. We evaluated diverse models including proprietary LLMs (GPT-"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2509.07135","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/2509.07135/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":"2026-07-05T12:07:16Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"484Ac2rMWoS+/VMNagblmWxoCaDn0XUErxjjzjnruMqs9u0vpPBY9eH9v/sXvZaIsiHJtzkDCod+MPWGP5llBA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-22T22:41:44.887451Z"},"content_sha256":"adb9635e852f25adad9f44e56d46fd1140ad98fd0b4997e68beb278bc5cff812","schema_version":"1.0","event_id":"sha256:adb9635e852f25adad9f44e56d46fd1140ad98fd0b4997e68beb278bc5cff812"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/CCAPLEAOYNSJ7KQRVR6VCZ6AJH/bundle.json","state_url":"https://pith.science/pith/CCAPLEAOYNSJ7KQRVR6VCZ6AJH/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/CCAPLEAOYNSJ7KQRVR6VCZ6AJH/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":"2026-08-22T22:41:44Z","links":{"resolver":"https://pith.science/pith/CCAPLEAOYNSJ7KQRVR6VCZ6AJH","bundle":"https://pith.science/pith/CCAPLEAOYNSJ7KQRVR6VCZ6AJH/bundle.json","state":"https://pith.science/pith/CCAPLEAOYNSJ7KQRVR6VCZ6AJH/state.json","well_known_bundle":"https://pith.science/.well-known/pith/CCAPLEAOYNSJ7KQRVR6VCZ6AJH/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:CCAPLEAOYNSJ7KQRVR6VCZ6AJH","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":"1ba31cd11022a75d96f4c6de560718852bab84f5d195bed1a4d34a0332aded85","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-09-08T18:39:35Z","title_canon_sha256":"db0abb7e10d7a9e6e277b22f0f208b71e8895d891280304cc78235675f33e297"},"schema_version":"1.0","source":{"id":"2509.07135","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2509.07135","created_at":"2026-07-05T12:07:16Z"},{"alias_kind":"arxiv_version","alias_value":"2509.07135v1","created_at":"2026-07-05T12:07:16Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2509.07135","created_at":"2026-07-05T12:07:16Z"},{"alias_kind":"pith_short_12","alias_value":"CCAPLEAOYNSJ","created_at":"2026-07-05T12:07:16Z"},{"alias_kind":"pith_short_16","alias_value":"CCAPLEAOYNSJ7KQR","created_at":"2026-07-05T12:07:16Z"},{"alias_kind":"pith_short_8","alias_value":"CCAPLEAO","created_at":"2026-07-05T12:07:16Z"}],"graph_snapshots":[{"event_id":"sha256:adb9635e852f25adad9f44e56d46fd1140ad98fd0b4997e68beb278bc5cff812","target":"graph","created_at":"2026-07-05T12:07:16Z","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/2509.07135/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Large language models (LLMs) show increasing potential in education, yet benchmarks for non-English languages in specialized domains remain scarce. We introduce MedBench-IT, the first comprehensive benchmark for evaluating LLMs on Italian medical university entrance examinations. Sourced from Edizioni Simone, a leading preparatory materials publisher, MedBench-IT comprises 17,410 expert-written multiple-choice questions across six subjects (Biology, Chemistry, Logic, General Culture, Mathematics, Physics) and three difficulty levels. We evaluated diverse models including proprietary LLMs (GPT-","authors_text":"Alessandro Angioi, Davide Sanna, Michelangelo Puliga, Roberto Marras, Ruggero Marino Lazzaroni","cross_cats":[],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-09-08T18:39:35Z","title":"MedBench-IT: A Comprehensive Benchmark for Evaluating Large Language Models on Italian Medical Entrance Examinations"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2509.07135","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:ef9a6ba4c61f5f12f5d545507524ecbe623271da58dbbf6c9b8cf2eeb8cc7cd3","target":"record","created_at":"2026-07-05T12:07:16Z","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":"1ba31cd11022a75d96f4c6de560718852bab84f5d195bed1a4d34a0332aded85","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-09-08T18:39:35Z","title_canon_sha256":"db0abb7e10d7a9e6e277b22f0f208b71e8895d891280304cc78235675f33e297"},"schema_version":"1.0","source":{"id":"2509.07135","kind":"arxiv","version":1}},"canonical_sha256":"1080f5900ec3649faa11ac7d5167c049f94b761828875ffd02aabf8ec6958aaa","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"1080f5900ec3649faa11ac7d5167c049f94b761828875ffd02aabf8ec6958aaa","first_computed_at":"2026-07-05T12:07:16.146914Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T12:07:16.146914Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"PD14APgaEamnRuSkGpPBcaednC9m4vILF1eX8uvvt7hUmEnNcSOgM3ohRPAYnZGppa6t341o7vo4TzSbGrhSBw==","signature_status":"signed_v1","signed_at":"2026-07-05T12:07:16.147429Z","signed_message":"canonical_sha256_bytes"},"source_id":"2509.07135","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:ef9a6ba4c61f5f12f5d545507524ecbe623271da58dbbf6c9b8cf2eeb8cc7cd3","sha256:adb9635e852f25adad9f44e56d46fd1140ad98fd0b4997e68beb278bc5cff812"],"state_sha256":"da04ff24784cfc188c0aa36590f46d408e940c99edc72c9291126da3829d6523"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"zjiKfIuBuHMWVew1BWxcqy1SaBZbQrQ7mUExAClaWxQ0ROdtpPmUrmnkw7wyDaY0uANKH6xDvIq2XL5ObgeUAg==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-22T22:41:44.895989Z","bundle_sha256":"77e1712b438160521a946ef9230934e3b5debcd8a5301bb6045f32182065db82"}}