{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:ZOLDDFDTQBJENJYZDGHWI2EGE2","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":"e4f7ed9918ab840fae6026ef35ec4bcb43042cae2af56a000e56f28533abab21","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-06-25T04:27:53Z","title_canon_sha256":"4956facb5ae58f3effd9547fccd8eec75d2064d4f2601e18b0dcdf4eef99a65e"},"schema_version":"1.0","source":{"id":"2406.17271","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2406.17271","created_at":"2026-07-05T08:36:32Z"},{"alias_kind":"arxiv_version","alias_value":"2406.17271v1","created_at":"2026-07-05T08:36:32Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2406.17271","created_at":"2026-07-05T08:36:32Z"},{"alias_kind":"pith_short_12","alias_value":"ZOLDDFDTQBJE","created_at":"2026-07-05T08:36:32Z"},{"alias_kind":"pith_short_16","alias_value":"ZOLDDFDTQBJENJYZ","created_at":"2026-07-05T08:36:32Z"},{"alias_kind":"pith_short_8","alias_value":"ZOLDDFDT","created_at":"2026-07-05T08:36:32Z"}],"graph_snapshots":[{"event_id":"sha256:a703970fe4c98e132f0da308d1bdd161055dbaf0d337ed5649cff8ae11174c16","target":"graph","created_at":"2026-07-05T08:36:32Z","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/2406.17271/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"The current paradigm of evaluating Large Language Models (LLMs) through static benchmarks comes with significant limitations, such as vulnerability to data contamination and a lack of adaptability to the evolving capabilities of LLMs. Therefore, evaluation methods that can adapt and generate evaluation data with controlled complexity are urgently needed. In this work, we introduce Dynamic Evaluation of LLMs via Adaptive Reasoning Graph Evolvement (DARG) to dynamically extend current benchmarks with controlled complexity and diversity. Specifically, we first extract the reasoning graphs of data","authors_text":"Diyi Yang, Jiaao Chen, Zhehao Zhang","cross_cats":[],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-06-25T04:27:53Z","title":"DARG: Dynamic Evaluation of Large Language Models via Adaptive Reasoning Graph"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2406.17271","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:f68ecc42d293bded7bd8dfc2df8174f4c7958b67b9836d5fb31f3521a7b5347b","target":"record","created_at":"2026-07-05T08:36:32Z","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":"e4f7ed9918ab840fae6026ef35ec4bcb43042cae2af56a000e56f28533abab21","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-06-25T04:27:53Z","title_canon_sha256":"4956facb5ae58f3effd9547fccd8eec75d2064d4f2601e18b0dcdf4eef99a65e"},"schema_version":"1.0","source":{"id":"2406.17271","kind":"arxiv","version":1}},"canonical_sha256":"cb96319473805246a719198f646886268580951e7747d7e9f5a839b25187fac7","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"cb96319473805246a719198f646886268580951e7747d7e9f5a839b25187fac7","first_computed_at":"2026-07-05T08:36:32.009302Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T08:36:32.009302Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"QKZ3bg9/zgML5jhctYU8+ODLWBEzuSvK/R93WU15Suj1iGRofV8sViy+7hNhwCLc7+O0IkQjksMzGku8Y1doAA==","signature_status":"signed_v1","signed_at":"2026-07-05T08:36:32.009649Z","signed_message":"canonical_sha256_bytes"},"source_id":"2406.17271","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:f68ecc42d293bded7bd8dfc2df8174f4c7958b67b9836d5fb31f3521a7b5347b","sha256:a703970fe4c98e132f0da308d1bdd161055dbaf0d337ed5649cff8ae11174c16"],"state_sha256":"e59d6792569f842fc832587f298fc21bbfe2ca8ac0ddd6f0a7cf3e158be6100f"}