{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2024:QNA3Z5KAG34JVCVARABYZBQKYR","short_pith_number":"pith:QNA3Z5KA","canonical_record":{"source":{"id":"2412.00539","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-11-30T17:09:49Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"3985a7d5e2b57a9a0a3eb9c5881b4041efc083a90be9835914bbd98a2555d43c","abstract_canon_sha256":"993439d385e9448ef5117890ac8662e8ce788913fae19239a9d4920231a67832"},"schema_version":"1.0"},"canonical_sha256":"8341bcf54036f89a8aa088038c860ac45bc564bfb592e11a6ad6215e433e6c04","source":{"kind":"arxiv","id":"2412.00539","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2412.00539","created_at":"2026-07-05T09:45:59Z"},{"alias_kind":"arxiv_version","alias_value":"2412.00539v2","created_at":"2026-07-05T09:45:59Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2412.00539","created_at":"2026-07-05T09:45:59Z"},{"alias_kind":"pith_short_12","alias_value":"QNA3Z5KAG34J","created_at":"2026-07-05T09:45:59Z"},{"alias_kind":"pith_short_16","alias_value":"QNA3Z5KAG34JVCVA","created_at":"2026-07-05T09:45:59Z"},{"alias_kind":"pith_short_8","alias_value":"QNA3Z5KA","created_at":"2026-07-05T09:45:59Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2024:QNA3Z5KAG34JVCVARABYZBQKYR","target":"record","payload":{"canonical_record":{"source":{"id":"2412.00539","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-11-30T17:09:49Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"3985a7d5e2b57a9a0a3eb9c5881b4041efc083a90be9835914bbd98a2555d43c","abstract_canon_sha256":"993439d385e9448ef5117890ac8662e8ce788913fae19239a9d4920231a67832"},"schema_version":"1.0"},"canonical_sha256":"8341bcf54036f89a8aa088038c860ac45bc564bfb592e11a6ad6215e433e6c04","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:45:59.103743Z","signature_b64":"Z/qGP7rKBOCbAsXYHBHq+C8/fRuzR8IH7DGe1K5oQyLAEyLlMDryKFUT/1PbzaE7gDEa5wiTgOg68D2PsRZ4DA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"8341bcf54036f89a8aa088038c860ac45bc564bfb592e11a6ad6215e433e6c04","last_reissued_at":"2026-07-05T09:45:59.103351Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:45:59.103351Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2412.00539","source_version":2,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T09:45:59Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"uAcYCSsUAwFN1iV6v5NVWxLx7HiJzxOAmvoJOs/AvvozAnwijEiPqwdk9oiHMuMknRjASXI5PKTUg1PYLqmEAA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-23T17:13:58.692625Z"},"content_sha256":"74cef9d0445d883bda5c36cd60c97e9b77e1515d143e1ef2ed68782343bb3622","schema_version":"1.0","event_id":"sha256:74cef9d0445d883bda5c36cd60c97e9b77e1515d143e1ef2ed68782343bb3622"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2024:QNA3Z5KAG34JVCVARABYZBQKYR","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"TextClass Benchmark: A Continuous Elo Rating of LLMs in Social Sciences","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Basti\\'an Gonz\\'alez-Bustamante","submitted_at":"2024-11-30T17:09:49Z","abstract_excerpt":"The TextClass Benchmark project is an ongoing, continuous benchmarking process that aims to provide a comprehensive, fair, and dynamic evaluation of LLMs and transformers for text classification tasks. This evaluation spans various domains and languages in social sciences disciplines engaged in NLP and text-as-data approach. The leaderboards present performance metrics and relative ranking using a tailored Elo rating system. With each leaderboard cycle, novel models are added, fixed test sets can be replaced for unseen, equivalent data to test generalisation power, ratings are updated, and a M"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2412.00539","kind":"arxiv","version":2},"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/2412.00539/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":"2026-07-05T09:45:59Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"SmLlXzBXQ6qemg/8VwrDj0sftVUcZJpEZUTE8dptpRW9UdtlcXPSR0DC8Pej0JocaiO8c2gFiFwhCMERhZ6+CA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-23T17:13:58.693123Z"},"content_sha256":"cb88191d3cb867571fd296400541bc474f87c942d9946714da68003a4dc699e5","schema_version":"1.0","event_id":"sha256:cb88191d3cb867571fd296400541bc474f87c942d9946714da68003a4dc699e5"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/QNA3Z5KAG34JVCVARABYZBQKYR/bundle.json","state_url":"https://pith.science/pith/QNA3Z5KAG34JVCVARABYZBQKYR/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/QNA3Z5KAG34JVCVARABYZBQKYR/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":"2026-08-23T17:13:58Z","links":{"resolver":"https://pith.science/pith/QNA3Z5KAG34JVCVARABYZBQKYR","bundle":"https://pith.science/pith/QNA3Z5KAG34JVCVARABYZBQKYR/bundle.json","state":"https://pith.science/pith/QNA3Z5KAG34JVCVARABYZBQKYR/state.json","well_known_bundle":"https://pith.science/.well-known/pith/QNA3Z5KAG34JVCVARABYZBQKYR/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:QNA3Z5KAG34JVCVARABYZBQKYR","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":"993439d385e9448ef5117890ac8662e8ce788913fae19239a9d4920231a67832","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-11-30T17:09:49Z","title_canon_sha256":"3985a7d5e2b57a9a0a3eb9c5881b4041efc083a90be9835914bbd98a2555d43c"},"schema_version":"1.0","source":{"id":"2412.00539","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2412.00539","created_at":"2026-07-05T09:45:59Z"},{"alias_kind":"arxiv_version","alias_value":"2412.00539v2","created_at":"2026-07-05T09:45:59Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2412.00539","created_at":"2026-07-05T09:45:59Z"},{"alias_kind":"pith_short_12","alias_value":"QNA3Z5KAG34J","created_at":"2026-07-05T09:45:59Z"},{"alias_kind":"pith_short_16","alias_value":"QNA3Z5KAG34JVCVA","created_at":"2026-07-05T09:45:59Z"},{"alias_kind":"pith_short_8","alias_value":"QNA3Z5KA","created_at":"2026-07-05T09:45:59Z"}],"graph_snapshots":[{"event_id":"sha256:cb88191d3cb867571fd296400541bc474f87c942d9946714da68003a4dc699e5","target":"graph","created_at":"2026-07-05T09:45:59Z","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/2412.00539/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"The TextClass Benchmark project is an ongoing, continuous benchmarking process that aims to provide a comprehensive, fair, and dynamic evaluation of LLMs and transformers for text classification tasks. This evaluation spans various domains and languages in social sciences disciplines engaged in NLP and text-as-data approach. The leaderboards present performance metrics and relative ranking using a tailored Elo rating system. With each leaderboard cycle, novel models are added, fixed test sets can be replaced for unseen, equivalent data to test generalisation power, ratings are updated, and a M","authors_text":"Basti\\'an Gonz\\'alez-Bustamante","cross_cats":["cs.AI"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-11-30T17:09:49Z","title":"TextClass Benchmark: A Continuous Elo Rating of LLMs in Social Sciences"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2412.00539","kind":"arxiv","version":2},"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":"sha256:74cef9d0445d883bda5c36cd60c97e9b77e1515d143e1ef2ed68782343bb3622","target":"record","created_at":"2026-07-05T09:45:59Z","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":"993439d385e9448ef5117890ac8662e8ce788913fae19239a9d4920231a67832","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-11-30T17:09:49Z","title_canon_sha256":"3985a7d5e2b57a9a0a3eb9c5881b4041efc083a90be9835914bbd98a2555d43c"},"schema_version":"1.0","source":{"id":"2412.00539","kind":"arxiv","version":2}},"canonical_sha256":"8341bcf54036f89a8aa088038c860ac45bc564bfb592e11a6ad6215e433e6c04","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"8341bcf54036f89a8aa088038c860ac45bc564bfb592e11a6ad6215e433e6c04","first_computed_at":"2026-07-05T09:45:59.103351Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T09:45:59.103351Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"Z/qGP7rKBOCbAsXYHBHq+C8/fRuzR8IH7DGe1K5oQyLAEyLlMDryKFUT/1PbzaE7gDEa5wiTgOg68D2PsRZ4DA==","signature_status":"signed_v1","signed_at":"2026-07-05T09:45:59.103743Z","signed_message":"canonical_sha256_bytes"},"source_id":"2412.00539","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:74cef9d0445d883bda5c36cd60c97e9b77e1515d143e1ef2ed68782343bb3622","sha256:cb88191d3cb867571fd296400541bc474f87c942d9946714da68003a4dc699e5"],"state_sha256":"31b3da5fb530caccef550171b598cae8ff3104ec97d8752e49dcb9f8dc80b589"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"nkS0xQLiM88uTFV3RdwZV+GJkIccgJYanF9oOo0uuyzdoI5sDGbTkGOzZruAQssXYjr8vMfdnh6D+RLTR0MQCw==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-23T17:13:58.696694Z","bundle_sha256":"787f311d37f9987607b086d06a369ffecc04c54e7b2be8347e1b7e6c283616c9"}}