{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2024:GBGRXUQNQ3RMSGSWEGVUV7HLPD","short_pith_number":"pith:GBGRXUQN","canonical_record":{"source":{"id":"2405.01535","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-05-02T17:59:35Z","cross_cats_sorted":[],"title_canon_sha256":"0e1c5998d6f70bfc614b0801c8de4ab32d013b641c65c7768656ca6124462267","abstract_canon_sha256":"7ca089558442566043a8874efa3ff54465ddb4068634b108c43fd812d5df5647"},"schema_version":"1.0"},"canonical_sha256":"304d1bd20d86e2c91a5621ab4afceb78f7305b0bc168fd54a015fb361d4831d0","source":{"kind":"arxiv","id":"2405.01535","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2405.01535","created_at":"2026-07-05T09:44:34Z"},{"alias_kind":"arxiv_version","alias_value":"2405.01535v2","created_at":"2026-07-05T09:44:34Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2405.01535","created_at":"2026-07-05T09:44:34Z"},{"alias_kind":"pith_short_12","alias_value":"GBGRXUQNQ3RM","created_at":"2026-07-05T09:44:34Z"},{"alias_kind":"pith_short_16","alias_value":"GBGRXUQNQ3RMSGSW","created_at":"2026-07-05T09:44:34Z"},{"alias_kind":"pith_short_8","alias_value":"GBGRXUQN","created_at":"2026-07-05T09:44:34Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2024:GBGRXUQNQ3RMSGSWEGVUV7HLPD","target":"record","payload":{"canonical_record":{"source":{"id":"2405.01535","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-05-02T17:59:35Z","cross_cats_sorted":[],"title_canon_sha256":"0e1c5998d6f70bfc614b0801c8de4ab32d013b641c65c7768656ca6124462267","abstract_canon_sha256":"7ca089558442566043a8874efa3ff54465ddb4068634b108c43fd812d5df5647"},"schema_version":"1.0"},"canonical_sha256":"304d1bd20d86e2c91a5621ab4afceb78f7305b0bc168fd54a015fb361d4831d0","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:44:34.247241Z","signature_b64":"eofJJksHaYBYmyoLLRZSv/aCigul9Lri6MPrOkZ2/+2Setp7aqreTKNaz1OVrfVwdjOm4vCSMBpwUqA5Vs6GDA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"304d1bd20d86e2c91a5621ab4afceb78f7305b0bc168fd54a015fb361d4831d0","last_reissued_at":"2026-07-05T09:44:34.246740Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:44:34.246740Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2405.01535","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":"2026-07-05T09:44:34Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"VXpxR0J6W0AFEF7sg5YEvkcDjC03qtP1Z2pw19kmqZI1zbGytjlSqP2j7RxVqXNLU0Z8uao1EE1ZCcD6RB9zDA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-18T02:53:17.130611Z"},"content_sha256":"3e0b3ea0dc974fca3c7dc544c2f8d4976d7a7720916abb1bc79b57051d37b7fb","schema_version":"1.0","event_id":"sha256:3e0b3ea0dc974fca3c7dc544c2f8d4976d7a7720916abb1bc79b57051d37b7fb"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2024:GBGRXUQNQ3RMSGSWEGVUV7HLPD","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Prometheus 2: An Open Source Language Model Specialized in Evaluating Other Language Models","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Bill Yuchen Lin, Graham Neubig, Jamin Shin, Juyoung Suk, Kyungjae Lee, Minjoon Seo, Moontae Lee, Sean Welleck, Seungone Kim, Shayne Longpre","submitted_at":"2024-05-02T17:59:35Z","abstract_excerpt":"Proprietary LMs such as GPT-4 are often employed to assess the quality of responses from various LMs. However, concerns including transparency, controllability, and affordability strongly motivate the development of open-source LMs specialized in evaluations. On the other hand, existing open evaluator LMs exhibit critical shortcomings: 1) they issue scores that significantly diverge from those assigned by humans, and 2) they lack the flexibility to perform both direct assessment and pairwise ranking, the two most prevalent forms of assessment. Additionally, they do not possess the ability to e"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2405.01535","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/2405.01535/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":"2026-07-05T09:44:34Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"9pVLzBQqbYarpV0Iq5DmdPE39J1AcyTY7+0lvXhSx1Ou9YVERVvd1lgBvDKfc2jr4qildrqeN6CH+S2hTO8BAg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-18T02:53:17.131151Z"},"content_sha256":"1d90727d6ff09a871a216456272935221452854bc12325aaa853a1dbddf621f9","schema_version":"1.0","event_id":"sha256:1d90727d6ff09a871a216456272935221452854bc12325aaa853a1dbddf621f9"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/GBGRXUQNQ3RMSGSWEGVUV7HLPD/bundle.json","state_url":"https://pith.science/pith/GBGRXUQNQ3RMSGSWEGVUV7HLPD/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/GBGRXUQNQ3RMSGSWEGVUV7HLPD/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":"2026-08-18T02:53:17Z","links":{"resolver":"https://pith.science/pith/GBGRXUQNQ3RMSGSWEGVUV7HLPD","bundle":"https://pith.science/pith/GBGRXUQNQ3RMSGSWEGVUV7HLPD/bundle.json","state":"https://pith.science/pith/GBGRXUQNQ3RMSGSWEGVUV7HLPD/state.json","well_known_bundle":"https://pith.science/.well-known/pith/GBGRXUQNQ3RMSGSWEGVUV7HLPD/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:GBGRXUQNQ3RMSGSWEGVUV7HLPD","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":"7ca089558442566043a8874efa3ff54465ddb4068634b108c43fd812d5df5647","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-05-02T17:59:35Z","title_canon_sha256":"0e1c5998d6f70bfc614b0801c8de4ab32d013b641c65c7768656ca6124462267"},"schema_version":"1.0","source":{"id":"2405.01535","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2405.01535","created_at":"2026-07-05T09:44:34Z"},{"alias_kind":"arxiv_version","alias_value":"2405.01535v2","created_at":"2026-07-05T09:44:34Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2405.01535","created_at":"2026-07-05T09:44:34Z"},{"alias_kind":"pith_short_12","alias_value":"GBGRXUQNQ3RM","created_at":"2026-07-05T09:44:34Z"},{"alias_kind":"pith_short_16","alias_value":"GBGRXUQNQ3RMSGSW","created_at":"2026-07-05T09:44:34Z"},{"alias_kind":"pith_short_8","alias_value":"GBGRXUQN","created_at":"2026-07-05T09:44:34Z"}],"graph_snapshots":[{"event_id":"sha256:1d90727d6ff09a871a216456272935221452854bc12325aaa853a1dbddf621f9","target":"graph","created_at":"2026-07-05T09:44:34Z","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/2405.01535/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Proprietary LMs such as GPT-4 are often employed to assess the quality of responses from various LMs. However, concerns including transparency, controllability, and affordability strongly motivate the development of open-source LMs specialized in evaluations. On the other hand, existing open evaluator LMs exhibit critical shortcomings: 1) they issue scores that significantly diverge from those assigned by humans, and 2) they lack the flexibility to perform both direct assessment and pairwise ranking, the two most prevalent forms of assessment. Additionally, they do not possess the ability to e","authors_text":"Bill Yuchen Lin, Graham Neubig, Jamin Shin, Juyoung Suk, Kyungjae Lee, Minjoon Seo, Moontae Lee, Sean Welleck, Seungone Kim, Shayne Longpre","cross_cats":[],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-05-02T17:59:35Z","title":"Prometheus 2: An Open Source Language Model Specialized in Evaluating Other Language Models"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2405.01535","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":"sha256:3e0b3ea0dc974fca3c7dc544c2f8d4976d7a7720916abb1bc79b57051d37b7fb","target":"record","created_at":"2026-07-05T09:44:34Z","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":"7ca089558442566043a8874efa3ff54465ddb4068634b108c43fd812d5df5647","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-05-02T17:59:35Z","title_canon_sha256":"0e1c5998d6f70bfc614b0801c8de4ab32d013b641c65c7768656ca6124462267"},"schema_version":"1.0","source":{"id":"2405.01535","kind":"arxiv","version":2}},"canonical_sha256":"304d1bd20d86e2c91a5621ab4afceb78f7305b0bc168fd54a015fb361d4831d0","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"304d1bd20d86e2c91a5621ab4afceb78f7305b0bc168fd54a015fb361d4831d0","first_computed_at":"2026-07-05T09:44:34.246740Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T09:44:34.246740Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"eofJJksHaYBYmyoLLRZSv/aCigul9Lri6MPrOkZ2/+2Setp7aqreTKNaz1OVrfVwdjOm4vCSMBpwUqA5Vs6GDA==","signature_status":"signed_v1","signed_at":"2026-07-05T09:44:34.247241Z","signed_message":"canonical_sha256_bytes"},"source_id":"2405.01535","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:3e0b3ea0dc974fca3c7dc544c2f8d4976d7a7720916abb1bc79b57051d37b7fb","sha256:1d90727d6ff09a871a216456272935221452854bc12325aaa853a1dbddf621f9"],"state_sha256":"6914e6c90c72fc64fc4bc818cba5868739907bbfe53298cf4e5a3ac8e8ee0a7a"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"MJ73kBrhCgEstLNWKzHkH3RNjdzCcyewSBSvF9Wbl68xAJTeYuOD+bW23fV/Xrj8vG+h0NrIoSWbMD/rdMnMCQ==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-18T02:53:17.134859Z","bundle_sha256":"743731819d42cf3f3e5ec8699ac7fec16f7f3d675a4ad47568fd8e8109106848"}}