{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2024:GMXIGQ7234KIE26LQY7EEDPCR2","short_pith_number":"pith:GMXIGQ72","canonical_record":{"source":{"id":"2411.07140","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2024-11-11T17:10:56Z","cross_cats_sorted":[],"title_canon_sha256":"a9f0f7460b974f1a47cfae96b30b95b8459aa13c552d2a8dcf919079fe506d03","abstract_canon_sha256":"b55fe4bb1c3e3ed58274ffd1f7c0b69fd53946303ccd30474f6d97a7a7ea6eba"},"schema_version":"1.0"},"canonical_sha256":"332e8343fadf14826bcb863e420de28e9855b5e4d27422df253c99b00cb5ec86","source":{"kind":"arxiv","id":"2411.07140","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2411.07140","created_at":"2026-07-05T09:34:58Z"},{"alias_kind":"arxiv_version","alias_value":"2411.07140v2","created_at":"2026-07-05T09:34:58Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2411.07140","created_at":"2026-07-05T09:34:58Z"},{"alias_kind":"pith_short_12","alias_value":"GMXIGQ7234KI","created_at":"2026-07-05T09:34:58Z"},{"alias_kind":"pith_short_16","alias_value":"GMXIGQ7234KIE26L","created_at":"2026-07-05T09:34:58Z"},{"alias_kind":"pith_short_8","alias_value":"GMXIGQ72","created_at":"2026-07-05T09:34:58Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2024:GMXIGQ7234KIE26LQY7EEDPCR2","target":"record","payload":{"canonical_record":{"source":{"id":"2411.07140","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2024-11-11T17:10:56Z","cross_cats_sorted":[],"title_canon_sha256":"a9f0f7460b974f1a47cfae96b30b95b8459aa13c552d2a8dcf919079fe506d03","abstract_canon_sha256":"b55fe4bb1c3e3ed58274ffd1f7c0b69fd53946303ccd30474f6d97a7a7ea6eba"},"schema_version":"1.0"},"canonical_sha256":"332e8343fadf14826bcb863e420de28e9855b5e4d27422df253c99b00cb5ec86","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:34:58.696056Z","signature_b64":"Rd8woI63ZxdHgmK6kSc1mCJbHThOe3s4qbyYfYUcSuzHnRI5FzvZyHzOM7TcsYSr6M9kxjip6+kxtPSatyI6BQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"332e8343fadf14826bcb863e420de28e9855b5e4d27422df253c99b00cb5ec86","last_reissued_at":"2026-07-05T09:34:58.695547Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:34:58.695547Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2411.07140","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":"2026-07-05T09:34:58Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"SnTKeLRsTnqp6cH+Yz9aCAJk2hwTP1SPGwYLLge3bdC5v9/UKVmwdSOalkO315b4/6zx8IU+5xBBQwtGQLgqDw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-25T16:31:03.520066Z"},"content_sha256":"dd5fcafa71db8a01f986fecec15d456aeb29fd9501968056fec7cfda4850132f","schema_version":"1.0","event_id":"sha256:dd5fcafa71db8a01f986fecec15d456aeb29fd9501968056fec7cfda4850132f"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2024:GMXIGQ7234KIE26LQY7EEDPCR2","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Chinese SimpleQA: A Chinese Factuality Evaluation for Large Language Models","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Boren Zheng, Bo Zheng, Chengwei Hu, Dekai Sun, Hangyu Guo, Hui Huang, Jiaheng Liu, Shilong Li, Shirong Lin, Weixun Wang, Wenbo Su, Xiaoyong Zhu, Xingyuan Bu, Xuepeng Liu, Yancheng He, Yingshui Tan, Zhicheng Zheng, Zhuoran Lin","submitted_at":"2024-11-11T17:10:56Z","abstract_excerpt":"New LLM evaluation benchmarks are important to align with the rapid development of Large Language Models (LLMs). In this work, we present Chinese SimpleQA, the first comprehensive Chinese benchmark to evaluate the factuality ability of language models to answer short questions, and Chinese SimpleQA mainly has five properties (i.e., Chinese, Diverse, High-quality, Static, Easy-to-evaluate). Specifically, first, we focus on the Chinese language over 6 major topics with 99 diverse subtopics. Second, we conduct a comprehensive quality control process to achieve high-quality questions and answers, "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2411.07140","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/2411.07140/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":"2026-07-05T09:34:58Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"lV7fwDdKedQ+l/wvEBikIhsW74X5Bf7j6meIdxlymNeW4EaZa4U8sYg/bFtLr3sCb165XXJH5vkYdLerW4VnDg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-25T16:31:03.520494Z"},"content_sha256":"c19fe67009a9931159922bc4195f1db1c91a63b4e60ddbc68ed443dc644ee93d","schema_version":"1.0","event_id":"sha256:c19fe67009a9931159922bc4195f1db1c91a63b4e60ddbc68ed443dc644ee93d"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/GMXIGQ7234KIE26LQY7EEDPCR2/bundle.json","state_url":"https://pith.science/pith/GMXIGQ7234KIE26LQY7EEDPCR2/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/GMXIGQ7234KIE26LQY7EEDPCR2/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":"2026-07-25T16:31:03Z","links":{"resolver":"https://pith.science/pith/GMXIGQ7234KIE26LQY7EEDPCR2","bundle":"https://pith.science/pith/GMXIGQ7234KIE26LQY7EEDPCR2/bundle.json","state":"https://pith.science/pith/GMXIGQ7234KIE26LQY7EEDPCR2/state.json","well_known_bundle":"https://pith.science/.well-known/pith/GMXIGQ7234KIE26LQY7EEDPCR2/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:GMXIGQ7234KIE26LQY7EEDPCR2","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":"b55fe4bb1c3e3ed58274ffd1f7c0b69fd53946303ccd30474f6d97a7a7ea6eba","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2024-11-11T17:10:56Z","title_canon_sha256":"a9f0f7460b974f1a47cfae96b30b95b8459aa13c552d2a8dcf919079fe506d03"},"schema_version":"1.0","source":{"id":"2411.07140","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2411.07140","created_at":"2026-07-05T09:34:58Z"},{"alias_kind":"arxiv_version","alias_value":"2411.07140v2","created_at":"2026-07-05T09:34:58Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2411.07140","created_at":"2026-07-05T09:34:58Z"},{"alias_kind":"pith_short_12","alias_value":"GMXIGQ7234KI","created_at":"2026-07-05T09:34:58Z"},{"alias_kind":"pith_short_16","alias_value":"GMXIGQ7234KIE26L","created_at":"2026-07-05T09:34:58Z"},{"alias_kind":"pith_short_8","alias_value":"GMXIGQ72","created_at":"2026-07-05T09:34:58Z"}],"graph_snapshots":[{"event_id":"sha256:c19fe67009a9931159922bc4195f1db1c91a63b4e60ddbc68ed443dc644ee93d","target":"graph","created_at":"2026-07-05T09:34:58Z","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/2411.07140/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"New LLM evaluation benchmarks are important to align with the rapid development of Large Language Models (LLMs). In this work, we present Chinese SimpleQA, the first comprehensive Chinese benchmark to evaluate the factuality ability of language models to answer short questions, and Chinese SimpleQA mainly has five properties (i.e., Chinese, Diverse, High-quality, Static, Easy-to-evaluate). Specifically, first, we focus on the Chinese language over 6 major topics with 99 diverse subtopics. Second, we conduct a comprehensive quality control process to achieve high-quality questions and answers, ","authors_text":"Boren Zheng, Bo Zheng, Chengwei Hu, Dekai Sun, Hangyu Guo, Hui Huang, Jiaheng Liu, Shilong Li, Shirong Lin, Weixun Wang, Wenbo Su, Xiaoyong Zhu, Xingyuan Bu, Xuepeng Liu, Yancheng He, Yingshui Tan, Zhicheng Zheng, Zhuoran Lin","cross_cats":[],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2024-11-11T17:10:56Z","title":"Chinese SimpleQA: A Chinese Factuality Evaluation for Large Language Models"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2411.07140","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:dd5fcafa71db8a01f986fecec15d456aeb29fd9501968056fec7cfda4850132f","target":"record","created_at":"2026-07-05T09:34:58Z","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":"b55fe4bb1c3e3ed58274ffd1f7c0b69fd53946303ccd30474f6d97a7a7ea6eba","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2024-11-11T17:10:56Z","title_canon_sha256":"a9f0f7460b974f1a47cfae96b30b95b8459aa13c552d2a8dcf919079fe506d03"},"schema_version":"1.0","source":{"id":"2411.07140","kind":"arxiv","version":2}},"canonical_sha256":"332e8343fadf14826bcb863e420de28e9855b5e4d27422df253c99b00cb5ec86","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"332e8343fadf14826bcb863e420de28e9855b5e4d27422df253c99b00cb5ec86","first_computed_at":"2026-07-05T09:34:58.695547Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T09:34:58.695547Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"Rd8woI63ZxdHgmK6kSc1mCJbHThOe3s4qbyYfYUcSuzHnRI5FzvZyHzOM7TcsYSr6M9kxjip6+kxtPSatyI6BQ==","signature_status":"signed_v1","signed_at":"2026-07-05T09:34:58.696056Z","signed_message":"canonical_sha256_bytes"},"source_id":"2411.07140","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:dd5fcafa71db8a01f986fecec15d456aeb29fd9501968056fec7cfda4850132f","sha256:c19fe67009a9931159922bc4195f1db1c91a63b4e60ddbc68ed443dc644ee93d"],"state_sha256":"cef5da8c2add865d3114e850554ea8eaa7c490e8c9c4ecbf7e63636180f527ae"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"1qRW/hLK2YemqbhPeE3UNSVDfUhrziEdAotdphI1WcDcTFn+71CWcMc7+hIvN8aM5NjhpsmvmlDJiwKJTbbaCg==","signed_message":"bundle_sha256_bytes","signed_at":"2026-07-25T16:31:03.523112Z","bundle_sha256":"3f030245920f5e81cbe25ca300d5c138f5592545eb12dddb55c15556199f29cf"}}