{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2023:SM2TNWV4GST3LIRQX3G4KCUMDG","short_pith_number":"pith:SM2TNWV4","canonical_record":{"source":{"id":"2307.03109","kind":"arxiv","version":9},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2023-07-06T16:28:35Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"ef701aee76467a8bf93611aa9ae1fcefb08f12032877534fcb4f5777d845ff4c","abstract_canon_sha256":"8d80b8b1b9f4e5bc9999b0afba0f055511feaaeca7c4157a544a597360bf7bba"},"schema_version":"1.0"},"canonical_sha256":"933536dabc34a7b5a230becdc50a8c19b41e4e09acfcc3af4587b90731990cd2","source":{"kind":"arxiv","id":"2307.03109","version":9},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2307.03109","created_at":"2026-07-05T07:28:44Z"},{"alias_kind":"arxiv_version","alias_value":"2307.03109v9","created_at":"2026-07-05T07:28:44Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2307.03109","created_at":"2026-07-05T07:28:44Z"},{"alias_kind":"pith_short_12","alias_value":"SM2TNWV4GST3","created_at":"2026-07-05T07:28:44Z"},{"alias_kind":"pith_short_16","alias_value":"SM2TNWV4GST3LIRQ","created_at":"2026-07-05T07:28:44Z"},{"alias_kind":"pith_short_8","alias_value":"SM2TNWV4","created_at":"2026-07-05T07:28:44Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2023:SM2TNWV4GST3LIRQX3G4KCUMDG","target":"record","payload":{"canonical_record":{"source":{"id":"2307.03109","kind":"arxiv","version":9},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2023-07-06T16:28:35Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"ef701aee76467a8bf93611aa9ae1fcefb08f12032877534fcb4f5777d845ff4c","abstract_canon_sha256":"8d80b8b1b9f4e5bc9999b0afba0f055511feaaeca7c4157a544a597360bf7bba"},"schema_version":"1.0"},"canonical_sha256":"933536dabc34a7b5a230becdc50a8c19b41e4e09acfcc3af4587b90731990cd2","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T07:28:44.966947Z","signature_b64":"pTSsqT3+ftl6q+MVCZxj/DO6uT6iJVv7yP9EcG7tw9Rh6cDhoTR9j1Rv3WjP/5SG6D66J2X0+lMDNkoRtIyxCg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"933536dabc34a7b5a230becdc50a8c19b41e4e09acfcc3af4587b90731990cd2","last_reissued_at":"2026-07-05T07:28:44.966451Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T07:28:44.966451Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2307.03109","source_version":9,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T07:28:44Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"HFRUoyVU5mrY1mszgpR0ly/bEbgzPm6KcukbGl8GPj1AzElVhY1baYLJ4O1fSSQBmIehEM5R1o/2zEwGxZCbCA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-31T21:51:46.936555Z"},"content_sha256":"d52bc7e323fedca7f44d25e5c06e2a0459490ba93d083648a6c56bdc21b7be3d","schema_version":"1.0","event_id":"sha256:d52bc7e323fedca7f44d25e5c06e2a0459490ba93d083648a6c56bdc21b7be3d"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2023:SM2TNWV4GST3LIRQX3G4KCUMDG","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"A Survey on Evaluation of Large Language Models","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Cunxiang Wang, Hao Chen, Jindong Wang, Kaijie Zhu, Linyi Yang, Philip S. Yu, Qiang Yang, Wei Ye, Xiaoyuan Yi, Xing Xie, Xu Wang, Yi Chang, Yidong Wang, Yuan Wu, Yue Zhang, Yupeng Chang","submitted_at":"2023-07-06T16:28:35Z","abstract_excerpt":"Large language models (LLMs) are gaining increasing popularity in both academia and industry, owing to their unprecedented performance in various applications. As LLMs continue to play a vital role in both research and daily use, their evaluation becomes increasingly critical, not only at the task level, but also at the society level for better understanding of their potential risks. Over the past years, significant efforts have been made to examine LLMs from various perspectives. This paper presents a comprehensive review of these evaluation methods for LLMs, focusing on three key dimensions:"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2307.03109","kind":"arxiv","version":9},"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/2307.03109/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":"2026-07-05T07:28:44Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"pR7f6nTyM9H5XTWmrhkSTpJfedE9m5RNjlDZ3u3PVc9nHuPZu4wCqVMFL1GD3LLZpULqviqzpAEl6XYYJx6rBA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-31T21:51:46.937102Z"},"content_sha256":"066be6538a57e05893e47829e609dc77ec717f532525d8ce91865a0861e97f04","schema_version":"1.0","event_id":"sha256:066be6538a57e05893e47829e609dc77ec717f532525d8ce91865a0861e97f04"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/SM2TNWV4GST3LIRQX3G4KCUMDG/bundle.json","state_url":"https://pith.science/pith/SM2TNWV4GST3LIRQX3G4KCUMDG/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/SM2TNWV4GST3LIRQX3G4KCUMDG/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":"2026-07-31T21:51:46Z","links":{"resolver":"https://pith.science/pith/SM2TNWV4GST3LIRQX3G4KCUMDG","bundle":"https://pith.science/pith/SM2TNWV4GST3LIRQX3G4KCUMDG/bundle.json","state":"https://pith.science/pith/SM2TNWV4GST3LIRQX3G4KCUMDG/state.json","well_known_bundle":"https://pith.science/.well-known/pith/SM2TNWV4GST3LIRQX3G4KCUMDG/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2023:SM2TNWV4GST3LIRQX3G4KCUMDG","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":"8d80b8b1b9f4e5bc9999b0afba0f055511feaaeca7c4157a544a597360bf7bba","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2023-07-06T16:28:35Z","title_canon_sha256":"ef701aee76467a8bf93611aa9ae1fcefb08f12032877534fcb4f5777d845ff4c"},"schema_version":"1.0","source":{"id":"2307.03109","kind":"arxiv","version":9}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2307.03109","created_at":"2026-07-05T07:28:44Z"},{"alias_kind":"arxiv_version","alias_value":"2307.03109v9","created_at":"2026-07-05T07:28:44Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2307.03109","created_at":"2026-07-05T07:28:44Z"},{"alias_kind":"pith_short_12","alias_value":"SM2TNWV4GST3","created_at":"2026-07-05T07:28:44Z"},{"alias_kind":"pith_short_16","alias_value":"SM2TNWV4GST3LIRQ","created_at":"2026-07-05T07:28:44Z"},{"alias_kind":"pith_short_8","alias_value":"SM2TNWV4","created_at":"2026-07-05T07:28:44Z"}],"graph_snapshots":[{"event_id":"sha256:066be6538a57e05893e47829e609dc77ec717f532525d8ce91865a0861e97f04","target":"graph","created_at":"2026-07-05T07:28:44Z","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/2307.03109/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Large language models (LLMs) are gaining increasing popularity in both academia and industry, owing to their unprecedented performance in various applications. As LLMs continue to play a vital role in both research and daily use, their evaluation becomes increasingly critical, not only at the task level, but also at the society level for better understanding of their potential risks. Over the past years, significant efforts have been made to examine LLMs from various perspectives. This paper presents a comprehensive review of these evaluation methods for LLMs, focusing on three key dimensions:","authors_text":"Cunxiang Wang, Hao Chen, Jindong Wang, Kaijie Zhu, Linyi Yang, Philip S. Yu, Qiang Yang, Wei Ye, Xiaoyuan Yi, Xing Xie, Xu Wang, Yi Chang, Yidong Wang, Yuan Wu, Yue Zhang, Yupeng Chang","cross_cats":["cs.AI"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2023-07-06T16:28:35Z","title":"A Survey on Evaluation of Large Language Models"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2307.03109","kind":"arxiv","version":9},"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:d52bc7e323fedca7f44d25e5c06e2a0459490ba93d083648a6c56bdc21b7be3d","target":"record","created_at":"2026-07-05T07:28:44Z","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":"8d80b8b1b9f4e5bc9999b0afba0f055511feaaeca7c4157a544a597360bf7bba","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2023-07-06T16:28:35Z","title_canon_sha256":"ef701aee76467a8bf93611aa9ae1fcefb08f12032877534fcb4f5777d845ff4c"},"schema_version":"1.0","source":{"id":"2307.03109","kind":"arxiv","version":9}},"canonical_sha256":"933536dabc34a7b5a230becdc50a8c19b41e4e09acfcc3af4587b90731990cd2","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"933536dabc34a7b5a230becdc50a8c19b41e4e09acfcc3af4587b90731990cd2","first_computed_at":"2026-07-05T07:28:44.966451Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T07:28:44.966451Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"pTSsqT3+ftl6q+MVCZxj/DO6uT6iJVv7yP9EcG7tw9Rh6cDhoTR9j1Rv3WjP/5SG6D66J2X0+lMDNkoRtIyxCg==","signature_status":"signed_v1","signed_at":"2026-07-05T07:28:44.966947Z","signed_message":"canonical_sha256_bytes"},"source_id":"2307.03109","source_kind":"arxiv","source_version":9}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:d52bc7e323fedca7f44d25e5c06e2a0459490ba93d083648a6c56bdc21b7be3d","sha256:066be6538a57e05893e47829e609dc77ec717f532525d8ce91865a0861e97f04"],"state_sha256":"628f263173db861bab29b1e5bb9e7161047677507fb00a608ea2ca3e75e6d84f"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"UC+i8XmRv5bnU43GdpgqJ1G/ixqr+ywkkrIa3uJFBr2M0Tk91IBdn7YDfZiJtEwWXDw/y7vK7JH72Dm5ofwpBQ==","signed_message":"bundle_sha256_bytes","signed_at":"2026-07-31T21:51:46.943861Z","bundle_sha256":"305ad61d77def44812549d988aee14771a1ae87bac26e624c4942f87d3e7952e"}}