{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2025:6ULYO7YOESXIVEV2OHCTP6GVN4","short_pith_number":"pith:6ULYO7YO","canonical_record":{"source":{"id":"2506.01920","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2025-06-02T17:39:50Z","cross_cats_sorted":[],"title_canon_sha256":"7c653bc21a314054dd37c5e6d5d40577b0f5e3560c84f092a9fd820ca79e1ccf","abstract_canon_sha256":"9148237a86c9321f65db18586efa55e3c31420945d8c9e5b317b241b26e42077"},"schema_version":"1.0"},"canonical_sha256":"f517877f0e24ae8a92ba71c537f8d56f11da9539f4a04818dba35f1bcdda3315","source":{"kind":"arxiv","id":"2506.01920","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2506.01920","created_at":"2026-07-05T11:14:21Z"},{"alias_kind":"arxiv_version","alias_value":"2506.01920v1","created_at":"2026-07-05T11:14:21Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2506.01920","created_at":"2026-07-05T11:14:21Z"},{"alias_kind":"pith_short_12","alias_value":"6ULYO7YOESXI","created_at":"2026-07-05T11:14:21Z"},{"alias_kind":"pith_short_16","alias_value":"6ULYO7YOESXIVEV2","created_at":"2026-07-05T11:14:21Z"},{"alias_kind":"pith_short_8","alias_value":"6ULYO7YO","created_at":"2026-07-05T11:14:21Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2025:6ULYO7YOESXIVEV2OHCTP6GVN4","target":"record","payload":{"canonical_record":{"source":{"id":"2506.01920","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2025-06-02T17:39:50Z","cross_cats_sorted":[],"title_canon_sha256":"7c653bc21a314054dd37c5e6d5d40577b0f5e3560c84f092a9fd820ca79e1ccf","abstract_canon_sha256":"9148237a86c9321f65db18586efa55e3c31420945d8c9e5b317b241b26e42077"},"schema_version":"1.0"},"canonical_sha256":"f517877f0e24ae8a92ba71c537f8d56f11da9539f4a04818dba35f1bcdda3315","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:14:21.433840Z","signature_b64":"LCv76MRBYP6AgtYrugU+fwR0hag2qVutE5GCBNDDhIH8PIINS1LM6qqB4ta6MZ6MjTe86ZHzQBdFrJt+neb7Dw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"f517877f0e24ae8a92ba71c537f8d56f11da9539f4a04818dba35f1bcdda3315","last_reissued_at":"2026-07-05T11:14:21.433392Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:14:21.433392Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2506.01920","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":"2026-07-05T11:14:21Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"1eKpoDFsRCgJQr/s3+WySrb5+PngrEe+XAFKFlNm+aqFFA8+K8eYs60xGqmBA6uDyuRECozrGUSV6w3JCSFqCA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-09T00:44:21.039343Z"},"content_sha256":"849ccf423e1c38f3ae0de19dcaa4039ac8b9dc9b28b229a11e52ad3146fb7d27","schema_version":"1.0","event_id":"sha256:849ccf423e1c38f3ae0de19dcaa4039ac8b9dc9b28b229a11e52ad3146fb7d27"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2025:6ULYO7YOESXIVEV2OHCTP6GVN4","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"From Guidelines to Practice: A New Paradigm for Arabic Language Model Evaluation","license":"http://creativecommons.org/licenses/by-sa/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Abdulrahman Al-Batati, Adel Ammar, Omer Nacar, Serry Sibaee, Wadii Boulila, Yasser Al-Habashi","submitted_at":"2025-06-02T17:39:50Z","abstract_excerpt":"This paper addresses critical gaps in Arabic language model evaluation by establishing comprehensive theoretical guidelines and introducing a novel evaluation framework. We first analyze existing Arabic evaluation datasets, identifying significant issues in linguistic accuracy, cultural alignment, and methodological rigor. To address these limitations in LLMs, we present the Arabic Depth Mini Dataset (ADMD), a carefully curated collection of 490 challenging questions spanning ten major domains (42 sub-domains, see Figure 1. Using ADMD, we evaluate five leading language models: GPT-4, Claude 3."},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2506.01920","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/2506.01920/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":"2026-07-05T11:14:21Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"FXhfEzsNpYmpkZy1N0ESHwZGiKDCpEdkszZjXKN/8Touumu3v2uvncSIVkA+2djqesYb16u+r1c0nmGO7bJzCw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-09T00:44:21.039860Z"},"content_sha256":"93176eb3eaf81fff8816ccd8833bab2c68f8519520843cfd64b25c45a810cdf5","schema_version":"1.0","event_id":"sha256:93176eb3eaf81fff8816ccd8833bab2c68f8519520843cfd64b25c45a810cdf5"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/6ULYO7YOESXIVEV2OHCTP6GVN4/bundle.json","state_url":"https://pith.science/pith/6ULYO7YOESXIVEV2OHCTP6GVN4/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/6ULYO7YOESXIVEV2OHCTP6GVN4/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":"2026-08-09T00:44:21Z","links":{"resolver":"https://pith.science/pith/6ULYO7YOESXIVEV2OHCTP6GVN4","bundle":"https://pith.science/pith/6ULYO7YOESXIVEV2OHCTP6GVN4/bundle.json","state":"https://pith.science/pith/6ULYO7YOESXIVEV2OHCTP6GVN4/state.json","well_known_bundle":"https://pith.science/.well-known/pith/6ULYO7YOESXIVEV2OHCTP6GVN4/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:6ULYO7YOESXIVEV2OHCTP6GVN4","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":"9148237a86c9321f65db18586efa55e3c31420945d8c9e5b317b241b26e42077","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2025-06-02T17:39:50Z","title_canon_sha256":"7c653bc21a314054dd37c5e6d5d40577b0f5e3560c84f092a9fd820ca79e1ccf"},"schema_version":"1.0","source":{"id":"2506.01920","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2506.01920","created_at":"2026-07-05T11:14:21Z"},{"alias_kind":"arxiv_version","alias_value":"2506.01920v1","created_at":"2026-07-05T11:14:21Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2506.01920","created_at":"2026-07-05T11:14:21Z"},{"alias_kind":"pith_short_12","alias_value":"6ULYO7YOESXI","created_at":"2026-07-05T11:14:21Z"},{"alias_kind":"pith_short_16","alias_value":"6ULYO7YOESXIVEV2","created_at":"2026-07-05T11:14:21Z"},{"alias_kind":"pith_short_8","alias_value":"6ULYO7YO","created_at":"2026-07-05T11:14:21Z"}],"graph_snapshots":[{"event_id":"sha256:93176eb3eaf81fff8816ccd8833bab2c68f8519520843cfd64b25c45a810cdf5","target":"graph","created_at":"2026-07-05T11:14:21Z","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/2506.01920/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"This paper addresses critical gaps in Arabic language model evaluation by establishing comprehensive theoretical guidelines and introducing a novel evaluation framework. We first analyze existing Arabic evaluation datasets, identifying significant issues in linguistic accuracy, cultural alignment, and methodological rigor. To address these limitations in LLMs, we present the Arabic Depth Mini Dataset (ADMD), a carefully curated collection of 490 challenging questions spanning ten major domains (42 sub-domains, see Figure 1. Using ADMD, we evaluate five leading language models: GPT-4, Claude 3.","authors_text":"Abdulrahman Al-Batati, Adel Ammar, Omer Nacar, Serry Sibaee, Wadii Boulila, Yasser Al-Habashi","cross_cats":[],"headline":"","license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2025-06-02T17:39:50Z","title":"From Guidelines to Practice: A New Paradigm for Arabic Language Model Evaluation"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2506.01920","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":"sha256:849ccf423e1c38f3ae0de19dcaa4039ac8b9dc9b28b229a11e52ad3146fb7d27","target":"record","created_at":"2026-07-05T11:14:21Z","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":"9148237a86c9321f65db18586efa55e3c31420945d8c9e5b317b241b26e42077","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2025-06-02T17:39:50Z","title_canon_sha256":"7c653bc21a314054dd37c5e6d5d40577b0f5e3560c84f092a9fd820ca79e1ccf"},"schema_version":"1.0","source":{"id":"2506.01920","kind":"arxiv","version":1}},"canonical_sha256":"f517877f0e24ae8a92ba71c537f8d56f11da9539f4a04818dba35f1bcdda3315","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"f517877f0e24ae8a92ba71c537f8d56f11da9539f4a04818dba35f1bcdda3315","first_computed_at":"2026-07-05T11:14:21.433392Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T11:14:21.433392Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"LCv76MRBYP6AgtYrugU+fwR0hag2qVutE5GCBNDDhIH8PIINS1LM6qqB4ta6MZ6MjTe86ZHzQBdFrJt+neb7Dw==","signature_status":"signed_v1","signed_at":"2026-07-05T11:14:21.433840Z","signed_message":"canonical_sha256_bytes"},"source_id":"2506.01920","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:849ccf423e1c38f3ae0de19dcaa4039ac8b9dc9b28b229a11e52ad3146fb7d27","sha256:93176eb3eaf81fff8816ccd8833bab2c68f8519520843cfd64b25c45a810cdf5"],"state_sha256":"85292d1b4ca94fd84486cbf8a44b6dc642f40d9f01fa6f262bfc6e61f56e9d50"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"U5Zi9tWssHpH7KAReVvseqGgW1xQSsC8y1JS5mkm4uB13bsGH/aKp3XTlDy1DdYlafGeoX7SdoFYWZlLkkSACQ==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-09T00:44:21.043091Z","bundle_sha256":"1b7a2e7ac953471b0f5341aaf5cc8972d028b355301f06a7d74eaf6f0910ed45"}}