{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:XW6RXAGDXU3EH3J7MWDM7BV5CU","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":"8ba3513beaf455179cc0269a7f296637e9803c06ebf692252322ef03ed8e7cca","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-12-15T21:22:46Z","title_canon_sha256":"a7d7b4630a8089c06c9a134623699a5edbd1feee91f34990ca9391a024f73fdf"},"schema_version":"1.0","source":{"id":"2412.11314","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2412.11314","created_at":"2026-07-05T09:49:30Z"},{"alias_kind":"arxiv_version","alias_value":"2412.11314v1","created_at":"2026-07-05T09:49:30Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2412.11314","created_at":"2026-07-05T09:49:30Z"},{"alias_kind":"pith_short_12","alias_value":"XW6RXAGDXU3E","created_at":"2026-07-05T09:49:30Z"},{"alias_kind":"pith_short_16","alias_value":"XW6RXAGDXU3EH3J7","created_at":"2026-07-05T09:49:30Z"},{"alias_kind":"pith_short_8","alias_value":"XW6RXAGD","created_at":"2026-07-05T09:49:30Z"}],"graph_snapshots":[{"event_id":"sha256:61214e68231ad879b77368b7865dd17b3d6ebb34a23db01ff0921f9740121ba1","target":"graph","created_at":"2026-07-05T09:49:30Z","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/2412.11314/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"The rapid advancement of natural language processing (NLP) technologies, such as instruction-tuned large language models (LLMs), urges the development of modern evaluation protocols with human and machine feedback. We introduce Evalica, an open-source toolkit that facilitates the creation of reliable and reproducible model leaderboards. This paper presents its design, evaluates its performance, and demonstrates its usability through its Web interface, command-line interface, and Python API.","authors_text":"Dmitry Ustalov","cross_cats":[],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-12-15T21:22:46Z","title":"Reliable, Reproducible, and Really Fast Leaderboards with Evalica"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2412.11314","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":"sha256:7a7d608c1fb50b2ed8308d6ef3b8870341fd32671f35860c37c9592a6625373a","target":"record","created_at":"2026-07-05T09:49:30Z","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":"8ba3513beaf455179cc0269a7f296637e9803c06ebf692252322ef03ed8e7cca","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-12-15T21:22:46Z","title_canon_sha256":"a7d7b4630a8089c06c9a134623699a5edbd1feee91f34990ca9391a024f73fdf"},"schema_version":"1.0","source":{"id":"2412.11314","kind":"arxiv","version":1}},"canonical_sha256":"bdbd1b80c3bd3643ed3f6586cf86bd1512f2a0d74f647ebfab8fc5a63fe60fc0","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"bdbd1b80c3bd3643ed3f6586cf86bd1512f2a0d74f647ebfab8fc5a63fe60fc0","first_computed_at":"2026-07-05T09:49:30.472359Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T09:49:30.472359Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"QpcXLxRodixwheX3hmnUDc+oYatSU8WVcjd+2ubetgEALHZJ+1uiQWdiQE2GbFy7qS1mE6UJq3/knF+YvqbKCw==","signature_status":"signed_v1","signed_at":"2026-07-05T09:49:30.472779Z","signed_message":"canonical_sha256_bytes"},"source_id":"2412.11314","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:7a7d608c1fb50b2ed8308d6ef3b8870341fd32671f35860c37c9592a6625373a","sha256:61214e68231ad879b77368b7865dd17b3d6ebb34a23db01ff0921f9740121ba1"],"state_sha256":"49dd4991279e42fa2822da371f95fcc06895130f61ed458055ecdf6c42e1e095"}