{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:J2VH6V3Z7V3D43VLYZLPWUZ3IU","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":"e354dc7b9d721fb45328e8f81c8f8097559311c2ba9d83496218762200621dd0","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2024-04-11T09:17:12Z","title_canon_sha256":"c6d47666547705be113c7319d678757705ae05eb66634ca75edb2f9e84dca040"},"schema_version":"1.0","source":{"id":"2404.07584","kind":"arxiv","version":3}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2404.07584","created_at":"2026-07-05T08:46:23Z"},{"alias_kind":"arxiv_version","alias_value":"2404.07584v3","created_at":"2026-07-05T08:46:23Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2404.07584","created_at":"2026-07-05T08:46:23Z"},{"alias_kind":"pith_short_12","alias_value":"J2VH6V3Z7V3D","created_at":"2026-07-05T08:46:23Z"},{"alias_kind":"pith_short_16","alias_value":"J2VH6V3Z7V3D43VL","created_at":"2026-07-05T08:46:23Z"},{"alias_kind":"pith_short_8","alias_value":"J2VH6V3Z","created_at":"2026-07-05T08:46:23Z"}],"graph_snapshots":[{"event_id":"sha256:5e3085eb9cccdb840e5ae52e5bc237971520309fddcf73d9227b7003a9c330f8","target":"graph","created_at":"2026-07-05T08:46:23Z","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/2404.07584/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Evaluation is pivotal for refining Large Language Models (LLMs), pinpointing their capabilities, and guiding enhancements. The rapid development of LLMs calls for a lightweight and easy-to-use framework for swift evaluation deployment. However, considering various implementation details, developing a comprehensive evaluation platform is never easy. Existing platforms are often complex and poorly modularized, hindering seamless incorporation into research workflows. This paper introduces UltraEval, a user-friendly evaluation framework characterized by its lightweight nature, comprehensiveness, ","authors_text":"Chaoqun He, Hanghao Wu, Jiajie Zhang, Jie Zhou, Maosong Sun, Renjie Luo, Shengding Hu, Xu Han, Yuanqian Zhao, Zhiyuan Liu","cross_cats":[],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2024-04-11T09:17:12Z","title":"UltraEval: A Lightweight Platform for Flexible and Comprehensive Evaluation for LLMs"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2404.07584","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":"sha256:22be5add60654ae49b6b8aac6dcbb97d03ad27f3c156f80b79b428687584ec77","target":"record","created_at":"2026-07-05T08:46:23Z","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":"e354dc7b9d721fb45328e8f81c8f8097559311c2ba9d83496218762200621dd0","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2024-04-11T09:17:12Z","title_canon_sha256":"c6d47666547705be113c7319d678757705ae05eb66634ca75edb2f9e84dca040"},"schema_version":"1.0","source":{"id":"2404.07584","kind":"arxiv","version":3}},"canonical_sha256":"4eaa7f5779fd763e6eabc656fb533b450b0863ef59f52825f16de32f1fbf59c9","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"4eaa7f5779fd763e6eabc656fb533b450b0863ef59f52825f16de32f1fbf59c9","first_computed_at":"2026-07-05T08:46:23.853820Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T08:46:23.853820Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"HgAkLFsN3LaiaTICZcgRjob8Dd8hZaihydnS3BIaW2z6nompetsZ2uyne+NtwLTg4fcxWXMQ+UtN5VWUogCbBg==","signature_status":"signed_v1","signed_at":"2026-07-05T08:46:23.854365Z","signed_message":"canonical_sha256_bytes"},"source_id":"2404.07584","source_kind":"arxiv","source_version":3}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:22be5add60654ae49b6b8aac6dcbb97d03ad27f3c156f80b79b428687584ec77","sha256:5e3085eb9cccdb840e5ae52e5bc237971520309fddcf73d9227b7003a9c330f8"],"state_sha256":"8821552a390d225129c3eccf366990869d51f32665993122fd48c3b7abf6ab6c"}