{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2023:NRRST6X2D6B7E67LVESXHB75MI","short_pith_number":"pith:NRRST6X2","canonical_record":{"source":{"id":"2306.05783","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2023-06-09T09:52:05Z","cross_cats_sorted":[],"title_canon_sha256":"159f57f6031871c172fc01e10ddbe0051721f560045f0c1ffa744b63d50cc3cc","abstract_canon_sha256":"a758a33c9171ea315c4b22973c23015a77b29ec11a63a48b6bcbf039aaf2491f"},"schema_version":"1.0"},"canonical_sha256":"6c6329fafa1f83f27beba9257387fd621686cb6d3a01f3dfbd1753b16ae48e6c","source":{"kind":"arxiv","id":"2306.05783","version":3},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2306.05783","created_at":"2026-07-05T07:54:31Z"},{"alias_kind":"arxiv_version","alias_value":"2306.05783v3","created_at":"2026-07-05T07:54:31Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2306.05783","created_at":"2026-07-05T07:54:31Z"},{"alias_kind":"pith_short_12","alias_value":"NRRST6X2D6B7","created_at":"2026-07-05T07:54:31Z"},{"alias_kind":"pith_short_16","alias_value":"NRRST6X2D6B7E67L","created_at":"2026-07-05T07:54:31Z"},{"alias_kind":"pith_short_8","alias_value":"NRRST6X2","created_at":"2026-07-05T07:54:31Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2023:NRRST6X2D6B7E67LVESXHB75MI","target":"record","payload":{"canonical_record":{"source":{"id":"2306.05783","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2023-06-09T09:52:05Z","cross_cats_sorted":[],"title_canon_sha256":"159f57f6031871c172fc01e10ddbe0051721f560045f0c1ffa744b63d50cc3cc","abstract_canon_sha256":"a758a33c9171ea315c4b22973c23015a77b29ec11a63a48b6bcbf039aaf2491f"},"schema_version":"1.0"},"canonical_sha256":"6c6329fafa1f83f27beba9257387fd621686cb6d3a01f3dfbd1753b16ae48e6c","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T07:54:31.158331Z","signature_b64":"PljDEAPeoLtTjrXZWuRLVWcYP3bBFTTemyO1mOR6hsh2NdbzHQRIRS3U3LIZ2CU/OhCGnkyO2Tz2upVNfDRUAg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"6c6329fafa1f83f27beba9257387fd621686cb6d3a01f3dfbd1753b16ae48e6c","last_reissued_at":"2026-07-05T07:54:31.157849Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T07:54:31.157849Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2306.05783","source_version":3,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T07:54:31Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"hRt++t0TaelPjF3S7Z2gGTGYt8j9S53T8FofBF5Ww/5UZmbM9wde2BR+KXrITarrH5UGNV7AH8FQxnrW2XSCAQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-31T17:20:50.371509Z"},"content_sha256":"83250f9252d7a2dffed0b8c9c8e3490883f8638db4a4693a3080c700c89bda1d","schema_version":"1.0","event_id":"sha256:83250f9252d7a2dffed0b8c9c8e3490883f8638db4a4693a3080c700c89bda1d"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2023:NRRST6X2D6B7E67LVESXHB75MI","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Xiezhi: An Ever-Updating Benchmark for Holistic Domain Knowledge Evaluation","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Haoning Ye, Hongwei Feng, Jianchen Wang, Jingping Liu, Lin Zhang, Qianyu He, Rui Xu, Shusen Wang, Sihang Jiang, Weiguo Zheng, Weijie Wu, Wenhao Huang, Xiaoxuan Zhu, Yanghua Xiao, Yixin Zhu, Zhouhong Gu, Zhuozhi Xiong, Zihan Li, Zili Wang","submitted_at":"2023-06-09T09:52:05Z","abstract_excerpt":"New Natural Langauge Process~(NLP) benchmarks are urgently needed to align with the rapid development of large language models (LLMs). We present Xiezhi, the most comprehensive evaluation suite designed to assess holistic domain knowledge. Xiezhi comprises multiple-choice questions across 516 diverse disciplines ranging from 13 different subjects with 249,587 questions and accompanied by Xiezhi-Specialty and Xiezhi-Interdiscipline, both with 15k questions. We conduct evaluation of the 47 cutting-edge LLMs on Xiezhi. Results indicate that LLMs exceed average performance of humans in science, en"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2306.05783","kind":"arxiv","version":3},"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/2306.05783/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":"2026-07-05T07:54:31Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"mI66m7FpYLs0A7Rs+BOdO+t2TsBvBUP+SsCyy9KfeTBMYkntnM+LFSXqNJZ/jyU+prNk6P1p1MvltAmReCXqDQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-31T17:20:50.371953Z"},"content_sha256":"947414d888c7ec43102d844a81a4799f7b432577c18f455749aada7f4927edfc","schema_version":"1.0","event_id":"sha256:947414d888c7ec43102d844a81a4799f7b432577c18f455749aada7f4927edfc"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/NRRST6X2D6B7E67LVESXHB75MI/bundle.json","state_url":"https://pith.science/pith/NRRST6X2D6B7E67LVESXHB75MI/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/NRRST6X2D6B7E67LVESXHB75MI/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":"2026-07-31T17:20:50Z","links":{"resolver":"https://pith.science/pith/NRRST6X2D6B7E67LVESXHB75MI","bundle":"https://pith.science/pith/NRRST6X2D6B7E67LVESXHB75MI/bundle.json","state":"https://pith.science/pith/NRRST6X2D6B7E67LVESXHB75MI/state.json","well_known_bundle":"https://pith.science/.well-known/pith/NRRST6X2D6B7E67LVESXHB75MI/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2023:NRRST6X2D6B7E67LVESXHB75MI","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":"a758a33c9171ea315c4b22973c23015a77b29ec11a63a48b6bcbf039aaf2491f","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2023-06-09T09:52:05Z","title_canon_sha256":"159f57f6031871c172fc01e10ddbe0051721f560045f0c1ffa744b63d50cc3cc"},"schema_version":"1.0","source":{"id":"2306.05783","kind":"arxiv","version":3}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2306.05783","created_at":"2026-07-05T07:54:31Z"},{"alias_kind":"arxiv_version","alias_value":"2306.05783v3","created_at":"2026-07-05T07:54:31Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2306.05783","created_at":"2026-07-05T07:54:31Z"},{"alias_kind":"pith_short_12","alias_value":"NRRST6X2D6B7","created_at":"2026-07-05T07:54:31Z"},{"alias_kind":"pith_short_16","alias_value":"NRRST6X2D6B7E67L","created_at":"2026-07-05T07:54:31Z"},{"alias_kind":"pith_short_8","alias_value":"NRRST6X2","created_at":"2026-07-05T07:54:31Z"}],"graph_snapshots":[{"event_id":"sha256:947414d888c7ec43102d844a81a4799f7b432577c18f455749aada7f4927edfc","target":"graph","created_at":"2026-07-05T07:54:31Z","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/2306.05783/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"New Natural Langauge Process~(NLP) benchmarks are urgently needed to align with the rapid development of large language models (LLMs). We present Xiezhi, the most comprehensive evaluation suite designed to assess holistic domain knowledge. Xiezhi comprises multiple-choice questions across 516 diverse disciplines ranging from 13 different subjects with 249,587 questions and accompanied by Xiezhi-Specialty and Xiezhi-Interdiscipline, both with 15k questions. We conduct evaluation of the 47 cutting-edge LLMs on Xiezhi. Results indicate that LLMs exceed average performance of humans in science, en","authors_text":"Haoning Ye, Hongwei Feng, Jianchen Wang, Jingping Liu, Lin Zhang, Qianyu He, Rui Xu, Shusen Wang, Sihang Jiang, Weiguo Zheng, Weijie Wu, Wenhao Huang, Xiaoxuan Zhu, Yanghua Xiao, Yixin Zhu, Zhouhong Gu, Zhuozhi Xiong, Zihan Li, Zili Wang","cross_cats":[],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2023-06-09T09:52:05Z","title":"Xiezhi: An Ever-Updating Benchmark for Holistic Domain Knowledge Evaluation"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2306.05783","kind":"arxiv","version":3},"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":"sha256:83250f9252d7a2dffed0b8c9c8e3490883f8638db4a4693a3080c700c89bda1d","target":"record","created_at":"2026-07-05T07:54:31Z","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":"a758a33c9171ea315c4b22973c23015a77b29ec11a63a48b6bcbf039aaf2491f","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2023-06-09T09:52:05Z","title_canon_sha256":"159f57f6031871c172fc01e10ddbe0051721f560045f0c1ffa744b63d50cc3cc"},"schema_version":"1.0","source":{"id":"2306.05783","kind":"arxiv","version":3}},"canonical_sha256":"6c6329fafa1f83f27beba9257387fd621686cb6d3a01f3dfbd1753b16ae48e6c","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"6c6329fafa1f83f27beba9257387fd621686cb6d3a01f3dfbd1753b16ae48e6c","first_computed_at":"2026-07-05T07:54:31.157849Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T07:54:31.157849Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"PljDEAPeoLtTjrXZWuRLVWcYP3bBFTTemyO1mOR6hsh2NdbzHQRIRS3U3LIZ2CU/OhCGnkyO2Tz2upVNfDRUAg==","signature_status":"signed_v1","signed_at":"2026-07-05T07:54:31.158331Z","signed_message":"canonical_sha256_bytes"},"source_id":"2306.05783","source_kind":"arxiv","source_version":3}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:83250f9252d7a2dffed0b8c9c8e3490883f8638db4a4693a3080c700c89bda1d","sha256:947414d888c7ec43102d844a81a4799f7b432577c18f455749aada7f4927edfc"],"state_sha256":"3337baacc7ee11ebc26ccc20e7737b6fdc039bcd0f9ea558a6ab5a4413166f4e"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"pWsxxdDVmd09LQWmGBJfWnFW4FA2jGnRJa+4zaflhOglYgnFHpmVH9yQY4EfEwCuDDdvoYP4jjecY45qlXafAg==","signed_message":"bundle_sha256_bytes","signed_at":"2026-07-31T17:20:50.374917Z","bundle_sha256":"d7fea100ce65a1f153f03a7fedf9d54d66e8046ab19853227bee1044a2f2ab4e"}}