{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:TBY2NLQG6K6EC3BPV74CZ4C6SO","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":"e05b5bd87d335ca703eebd6fcc10f825a66fa3cbd90f8f7f5c3f1fee5bec4f3f","cross_cats_sorted":["cs.LG"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2025-05-19T07:34:25Z","title_canon_sha256":"2e425f957e6ff96409a0f4b7621844c4cebcf9db84dd5a5bbfceecef476df0fe"},"schema_version":"1.0","source":{"id":"2505.12808","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2505.12808","created_at":"2026-07-05T11:05:10Z"},{"alias_kind":"arxiv_version","alias_value":"2505.12808v1","created_at":"2026-07-05T11:05:10Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2505.12808","created_at":"2026-07-05T11:05:10Z"},{"alias_kind":"pith_short_12","alias_value":"TBY2NLQG6K6E","created_at":"2026-07-05T11:05:10Z"},{"alias_kind":"pith_short_16","alias_value":"TBY2NLQG6K6EC3BP","created_at":"2026-07-05T11:05:10Z"},{"alias_kind":"pith_short_8","alias_value":"TBY2NLQG","created_at":"2026-07-05T11:05:10Z"}],"graph_snapshots":[{"event_id":"sha256:9e9b04b05acd770b903145f0df8f5a5a68f528769fe62d9ad6fa0fa719194a2e","target":"graph","created_at":"2026-07-05T11:05:10Z","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/2505.12808/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"The recent explosion of large language models (LLMs), each with its own general or specialized strengths, makes scalable, reliable benchmarking more urgent than ever. Standard practices nowadays face fundamental trade-offs: closed-ended question-based benchmarks (eg MMLU) struggle with saturation as newer models emerge, while crowd-sourced leaderboards (eg Chatbot Arena) rely on costly and slow human judges. Recently, automated methods (eg LLM-as-a-judge) shed light on the scalability, but risk bias by relying on one or a few \"authority\" models. To tackle these issues, we propose Decentralized","authors_text":"Eric P. Xing, Haojian Jin, Jieyuan Liu, Kun Zhou, Shibo Hao, Somanshu Singla, Tianyang Liu, Xiangdong Zhang, Yanbin Yin, Yifei Shao, Yi Gu, Zhengzhong Liu, Zhen Wang, Zhiting Hu","cross_cats":["cs.LG"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2025-05-19T07:34:25Z","title":"Decentralized Arena: Towards Democratic and Scalable Automatic Evaluation of Language Models"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2505.12808","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:ea7191467a57c9c003326fb5dade24c5b7c36a41b175231609058b16bd1a2ad1","target":"record","created_at":"2026-07-05T11:05:10Z","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":"e05b5bd87d335ca703eebd6fcc10f825a66fa3cbd90f8f7f5c3f1fee5bec4f3f","cross_cats_sorted":["cs.LG"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2025-05-19T07:34:25Z","title_canon_sha256":"2e425f957e6ff96409a0f4b7621844c4cebcf9db84dd5a5bbfceecef476df0fe"},"schema_version":"1.0","source":{"id":"2505.12808","kind":"arxiv","version":1}},"canonical_sha256":"9871a6ae06f2bc416c2faff82cf05e939ddb6fc0c31bd9aa3da36195495dfae0","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"9871a6ae06f2bc416c2faff82cf05e939ddb6fc0c31bd9aa3da36195495dfae0","first_computed_at":"2026-07-05T11:05:10.039533Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T11:05:10.039533Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"i0raXnMzNHruJwZA85XBfn+FmJO5I2jZWDv50bqva8nZap1JBNMM0c6DtsrwcK8pa5G32yMYKAHZ07nP6jpdAQ==","signature_status":"signed_v1","signed_at":"2026-07-05T11:05:10.039976Z","signed_message":"canonical_sha256_bytes"},"source_id":"2505.12808","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:ea7191467a57c9c003326fb5dade24c5b7c36a41b175231609058b16bd1a2ad1","sha256:9e9b04b05acd770b903145f0df8f5a5a68f528769fe62d9ad6fa0fa719194a2e"],"state_sha256":"9e03b77e94236a1a4c000f9ffb1fb1fbaff45d17b801c76edb9a3fcad851c96a"}