{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2026:J4DCT3NR6EMUBKM5FBGIQAD4FI","short_pith_number":"pith:J4DCT3NR","canonical_record":{"source":{"id":"2605.21748","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2026-05-20T21:20:01Z","cross_cats_sorted":[],"title_canon_sha256":"44d1c390b8e275859c835f357326ea5933ced86b1fda3b669302f48540385bca","abstract_canon_sha256":"5c30d7991a03fd545f9e4342e235490c765c95cc1cacb8c41eb5da2ae1ec947f"},"schema_version":"1.0"},"canonical_sha256":"4f0629edb1f11940a99d284c88007c2a365d5832cb1f3804f09ca5489b724df7","source":{"kind":"arxiv","id":"2605.21748","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2605.21748","created_at":"2026-05-22T01:03:30Z"},{"alias_kind":"arxiv_version","alias_value":"2605.21748v1","created_at":"2026-05-22T01:03:30Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.21748","created_at":"2026-05-22T01:03:30Z"},{"alias_kind":"pith_short_12","alias_value":"J4DCT3NR6EMU","created_at":"2026-05-22T01:03:30Z"},{"alias_kind":"pith_short_16","alias_value":"J4DCT3NR6EMUBKM5","created_at":"2026-05-22T01:03:30Z"},{"alias_kind":"pith_short_8","alias_value":"J4DCT3NR","created_at":"2026-05-22T01:03:30Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2026:J4DCT3NR6EMUBKM5FBGIQAD4FI","target":"record","payload":{"canonical_record":{"source":{"id":"2605.21748","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2026-05-20T21:20:01Z","cross_cats_sorted":[],"title_canon_sha256":"44d1c390b8e275859c835f357326ea5933ced86b1fda3b669302f48540385bca","abstract_canon_sha256":"5c30d7991a03fd545f9e4342e235490c765c95cc1cacb8c41eb5da2ae1ec947f"},"schema_version":"1.0"},"canonical_sha256":"4f0629edb1f11940a99d284c88007c2a365d5832cb1f3804f09ca5489b724df7","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-22T01:03:30.415150Z","signature_b64":"rFdp7uCpmF1uCXyvLyrL3X28fOoHqBk32AaN96KKoHBGkUNV8/8YW4zix6dfDc2nOJuGF2HBk+EmXzGyLjpUAQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"4f0629edb1f11940a99d284c88007c2a365d5832cb1f3804f09ca5489b724df7","last_reissued_at":"2026-05-22T01:03:30.414709Z","signature_status":"signed_v1","first_computed_at":"2026-05-22T01:03:30.414709Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2605.21748","source_version":1,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-05-22T01:03:30Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"dd6ZrXbkO9uXIcmt/8Q1YZd2JlULXq9PPjl8tC7hyKuKMK0qxia3tlLN7MXCRYz+krwKrqsvm2xPwGE8SMXBBg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-23T07:32:09.424563Z"},"content_sha256":"4686cb5bfd73b8d7a48850aaa0f254e71814c66b6526076674e52a03ee337b62","schema_version":"1.0","event_id":"sha256:4686cb5bfd73b8d7a48850aaa0f254e71814c66b6526076674e52a03ee337b62"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2026:J4DCT3NR6EMUBKM5FBGIQAD4FI","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"RankJudge: A Multi-Turn LLM-as-a-Judge Synthetic Benchmark Generator","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Jesse C. Cresswell, Keyvan Golestan, Rasa Hosseinzadeh, Tongzi Wu, Zhaoyan Liu, Zhenwei Tang","submitted_at":"2026-05-20T21:20:01Z","abstract_excerpt":"As interactive LLM-based applications are created and refined, model developers need to evaluate the quality of generated text along many possible axes. For simpler systems, human evaluation may be practical, but in complicated systems like conversational chatbots, the amount of generated text can overwhelm human annotation resources. Model developers have begun to rely heavily on auto-evaluation, where LLMs are also used to judge generation quality. However, existing LLM-as-a-judge benchmarks largely focus on simple Q\\&A tasks that do not match the complexity of multi-turn conversations. We i"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2605.21748","kind":"arxiv","version":1},"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/2605.21748/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":"2026-05-22T01:03:30Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"Y3sq7G1XM0L/Xs+BcAGTDoOdrsEtfS0TUSmiIozmRacgNrIRFDEzJHB9eyy/dl0W94lkcGJRxHGJuQ7ynPn+Dg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-23T07:32:09.425250Z"},"content_sha256":"40b06f736c90231237995ab70fa4baa4a5c1acf03bb72062b913660554bd1d2f","schema_version":"1.0","event_id":"sha256:40b06f736c90231237995ab70fa4baa4a5c1acf03bb72062b913660554bd1d2f"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/J4DCT3NR6EMUBKM5FBGIQAD4FI/bundle.json","state_url":"https://pith.science/pith/J4DCT3NR6EMUBKM5FBGIQAD4FI/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/J4DCT3NR6EMUBKM5FBGIQAD4FI/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":"2026-05-23T07:32:09Z","links":{"resolver":"https://pith.science/pith/J4DCT3NR6EMUBKM5FBGIQAD4FI","bundle":"https://pith.science/pith/J4DCT3NR6EMUBKM5FBGIQAD4FI/bundle.json","state":"https://pith.science/pith/J4DCT3NR6EMUBKM5FBGIQAD4FI/state.json","well_known_bundle":"https://pith.science/.well-known/pith/J4DCT3NR6EMUBKM5FBGIQAD4FI/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:J4DCT3NR6EMUBKM5FBGIQAD4FI","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":"5c30d7991a03fd545f9e4342e235490c765c95cc1cacb8c41eb5da2ae1ec947f","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2026-05-20T21:20:01Z","title_canon_sha256":"44d1c390b8e275859c835f357326ea5933ced86b1fda3b669302f48540385bca"},"schema_version":"1.0","source":{"id":"2605.21748","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2605.21748","created_at":"2026-05-22T01:03:30Z"},{"alias_kind":"arxiv_version","alias_value":"2605.21748v1","created_at":"2026-05-22T01:03:30Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.21748","created_at":"2026-05-22T01:03:30Z"},{"alias_kind":"pith_short_12","alias_value":"J4DCT3NR6EMU","created_at":"2026-05-22T01:03:30Z"},{"alias_kind":"pith_short_16","alias_value":"J4DCT3NR6EMUBKM5","created_at":"2026-05-22T01:03:30Z"},{"alias_kind":"pith_short_8","alias_value":"J4DCT3NR","created_at":"2026-05-22T01:03:30Z"}],"graph_snapshots":[{"event_id":"sha256:40b06f736c90231237995ab70fa4baa4a5c1acf03bb72062b913660554bd1d2f","target":"graph","created_at":"2026-05-22T01:03:30Z","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/2605.21748/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"As interactive LLM-based applications are created and refined, model developers need to evaluate the quality of generated text along many possible axes. For simpler systems, human evaluation may be practical, but in complicated systems like conversational chatbots, the amount of generated text can overwhelm human annotation resources. Model developers have begun to rely heavily on auto-evaluation, where LLMs are also used to judge generation quality. However, existing LLM-as-a-judge benchmarks largely focus on simple Q\\&A tasks that do not match the complexity of multi-turn conversations. We i","authors_text":"Jesse C. Cresswell, Keyvan Golestan, Rasa Hosseinzadeh, Tongzi Wu, Zhaoyan Liu, Zhenwei Tang","cross_cats":[],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2026-05-20T21:20:01Z","title":"RankJudge: A Multi-Turn LLM-as-a-Judge Synthetic Benchmark Generator"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2605.21748","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":"sha256:4686cb5bfd73b8d7a48850aaa0f254e71814c66b6526076674e52a03ee337b62","target":"record","created_at":"2026-05-22T01:03:30Z","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":"5c30d7991a03fd545f9e4342e235490c765c95cc1cacb8c41eb5da2ae1ec947f","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2026-05-20T21:20:01Z","title_canon_sha256":"44d1c390b8e275859c835f357326ea5933ced86b1fda3b669302f48540385bca"},"schema_version":"1.0","source":{"id":"2605.21748","kind":"arxiv","version":1}},"canonical_sha256":"4f0629edb1f11940a99d284c88007c2a365d5832cb1f3804f09ca5489b724df7","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"4f0629edb1f11940a99d284c88007c2a365d5832cb1f3804f09ca5489b724df7","first_computed_at":"2026-05-22T01:03:30.414709Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-22T01:03:30.414709Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"rFdp7uCpmF1uCXyvLyrL3X28fOoHqBk32AaN96KKoHBGkUNV8/8YW4zix6dfDc2nOJuGF2HBk+EmXzGyLjpUAQ==","signature_status":"signed_v1","signed_at":"2026-05-22T01:03:30.415150Z","signed_message":"canonical_sha256_bytes"},"source_id":"2605.21748","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:4686cb5bfd73b8d7a48850aaa0f254e71814c66b6526076674e52a03ee337b62","sha256:40b06f736c90231237995ab70fa4baa4a5c1acf03bb72062b913660554bd1d2f"],"state_sha256":"1e569ae5c9b3bc7be33154ad1bc112fdc0b085098c8fbd8e98b74ff3ef9f0a2c"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"HS9gHWwo0j3PujCi/oTHm4DxGYEGm2cXw5eLZOCf6pWQ0QYhtOj6qeJXf0qx+laaIHpqbp9nCTpkZeagk/JSCw==","signed_message":"bundle_sha256_bytes","signed_at":"2026-05-23T07:32:09.428166Z","bundle_sha256":"8970ef9278fae2dd8345b6f93e70e7a268052d91d721a62996d7c18f461cb0ff"}}