{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2026:4AME2ISUSHHU4CFR6J772F4I4N","short_pith_number":"pith:4AME2ISU","canonical_record":{"source":{"id":"2606.17634","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2026-06-16T07:44:45Z","cross_cats_sorted":[],"title_canon_sha256":"fdc639872f881e0da22985ba983b84461cb9dec3fc192cfa5d214b875a4b6314","abstract_canon_sha256":"35ee847229f8423aef375face6d67886963aa2650abd191e2a0ee6624607a965"},"schema_version":"1.0"},"canonical_sha256":"e0184d225491cf4e08b1f27ffd1788e3799c367d187f416820e585ce23a603f3","source":{"kind":"arxiv","id":"2606.17634","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2606.17634","created_at":"2026-06-19T16:10:17Z"},{"alias_kind":"arxiv_version","alias_value":"2606.17634v1","created_at":"2026-06-19T16:10:17Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2606.17634","created_at":"2026-06-19T16:10:17Z"},{"alias_kind":"pith_short_12","alias_value":"4AME2ISUSHHU","created_at":"2026-06-19T16:10:17Z"},{"alias_kind":"pith_short_16","alias_value":"4AME2ISUSHHU4CFR","created_at":"2026-06-19T16:10:17Z"},{"alias_kind":"pith_short_8","alias_value":"4AME2ISU","created_at":"2026-06-19T16:10:17Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2026:4AME2ISUSHHU4CFR6J772F4I4N","target":"record","payload":{"canonical_record":{"source":{"id":"2606.17634","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2026-06-16T07:44:45Z","cross_cats_sorted":[],"title_canon_sha256":"fdc639872f881e0da22985ba983b84461cb9dec3fc192cfa5d214b875a4b6314","abstract_canon_sha256":"35ee847229f8423aef375face6d67886963aa2650abd191e2a0ee6624607a965"},"schema_version":"1.0"},"canonical_sha256":"e0184d225491cf4e08b1f27ffd1788e3799c367d187f416820e585ce23a603f3","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-19T16:10:17.780044Z","signature_b64":"0GDnerp5+uVzTXHXEFyxqGn18iIeXaIM61Ro31ssuBdAO4M/znJozeK7YwzLv58aMYX0THPkoIn0vZI8EkLzCw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"e0184d225491cf4e08b1f27ffd1788e3799c367d187f416820e585ce23a603f3","last_reissued_at":"2026-06-19T16:10:17.779705Z","signature_status":"signed_v1","first_computed_at":"2026-06-19T16:10:17.779705Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2606.17634","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":"2026-06-19T16:10:17Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"gY5ob0b2byTfS9v2kswb1EhTMEepwN3LV/Q/XnLrqLzbn+N+iQ4yIaOQoMSOaCRWbg09dZDt+NXFDfo7kilvCQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-04T15:27:35.014614Z"},"content_sha256":"d433a1eef87678e8ab57d118ffb6da749acdeaf4b913fbd65101d3632b189508","schema_version":"1.0","event_id":"sha256:d433a1eef87678e8ab57d118ffb6da749acdeaf4b913fbd65101d3632b189508"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2026:4AME2ISUSHHU4CFR6J772F4I4N","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Prompt Perturbation for Reliable LLM Evaluation over Comparison Graphs","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Dong Huang, Jianbo Sun, Pengkun Yang","submitted_at":"2026-06-16T07:44:45Z","abstract_excerpt":"Evaluating large language models (LLMs) is important for understanding their capabilities, comparing competing systems, and supporting the deployment of reliable models in practice. For open-ended tasks, pairwise evaluation has become a popular paradigm, in which two responses to the same prompt are compared and the resulting judgments are aggregated into an overall ranking. A central challenge of this paradigm is intransitivity: the induced comparison outcomes may fail to support any coherent global ranking. For example, one may observe cyclic preferences such as $A \\succ B \\succ C \\succ A$, "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2606.17634","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/2606.17634/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":"2026-06-19T16:10:17Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"jVqfpkBWs4UdJKfK3ovrZPsfD4qPnHvd732/huuVCw7dMUvSgFPLhnWFfDygbUebY69HmEimGnFDY2DjF5NOCg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-04T15:27:35.014999Z"},"content_sha256":"2dd695352abc56413ad5ee932b3d26b7aa34368e7592d8fe396af4c9fc0a68d5","schema_version":"1.0","event_id":"sha256:2dd695352abc56413ad5ee932b3d26b7aa34368e7592d8fe396af4c9fc0a68d5"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/4AME2ISUSHHU4CFR6J772F4I4N/bundle.json","state_url":"https://pith.science/pith/4AME2ISUSHHU4CFR6J772F4I4N/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/4AME2ISUSHHU4CFR6J772F4I4N/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":"2026-07-04T15:27:35Z","links":{"resolver":"https://pith.science/pith/4AME2ISUSHHU4CFR6J772F4I4N","bundle":"https://pith.science/pith/4AME2ISUSHHU4CFR6J772F4I4N/bundle.json","state":"https://pith.science/pith/4AME2ISUSHHU4CFR6J772F4I4N/state.json","well_known_bundle":"https://pith.science/.well-known/pith/4AME2ISUSHHU4CFR6J772F4I4N/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:4AME2ISUSHHU4CFR6J772F4I4N","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":"35ee847229f8423aef375face6d67886963aa2650abd191e2a0ee6624607a965","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2026-06-16T07:44:45Z","title_canon_sha256":"fdc639872f881e0da22985ba983b84461cb9dec3fc192cfa5d214b875a4b6314"},"schema_version":"1.0","source":{"id":"2606.17634","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2606.17634","created_at":"2026-06-19T16:10:17Z"},{"alias_kind":"arxiv_version","alias_value":"2606.17634v1","created_at":"2026-06-19T16:10:17Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2606.17634","created_at":"2026-06-19T16:10:17Z"},{"alias_kind":"pith_short_12","alias_value":"4AME2ISUSHHU","created_at":"2026-06-19T16:10:17Z"},{"alias_kind":"pith_short_16","alias_value":"4AME2ISUSHHU4CFR","created_at":"2026-06-19T16:10:17Z"},{"alias_kind":"pith_short_8","alias_value":"4AME2ISU","created_at":"2026-06-19T16:10:17Z"}],"graph_snapshots":[{"event_id":"sha256:2dd695352abc56413ad5ee932b3d26b7aa34368e7592d8fe396af4c9fc0a68d5","target":"graph","created_at":"2026-06-19T16:10:17Z","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/2606.17634/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Evaluating large language models (LLMs) is important for understanding their capabilities, comparing competing systems, and supporting the deployment of reliable models in practice. For open-ended tasks, pairwise evaluation has become a popular paradigm, in which two responses to the same prompt are compared and the resulting judgments are aggregated into an overall ranking. A central challenge of this paradigm is intransitivity: the induced comparison outcomes may fail to support any coherent global ranking. For example, one may observe cyclic preferences such as $A \\succ B \\succ C \\succ A$, ","authors_text":"Dong Huang, Jianbo Sun, Pengkun Yang","cross_cats":[],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2026-06-16T07:44:45Z","title":"Prompt Perturbation for Reliable LLM Evaluation over Comparison Graphs"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2606.17634","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:d433a1eef87678e8ab57d118ffb6da749acdeaf4b913fbd65101d3632b189508","target":"record","created_at":"2026-06-19T16:10:17Z","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":"35ee847229f8423aef375face6d67886963aa2650abd191e2a0ee6624607a965","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2026-06-16T07:44:45Z","title_canon_sha256":"fdc639872f881e0da22985ba983b84461cb9dec3fc192cfa5d214b875a4b6314"},"schema_version":"1.0","source":{"id":"2606.17634","kind":"arxiv","version":1}},"canonical_sha256":"e0184d225491cf4e08b1f27ffd1788e3799c367d187f416820e585ce23a603f3","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"e0184d225491cf4e08b1f27ffd1788e3799c367d187f416820e585ce23a603f3","first_computed_at":"2026-06-19T16:10:17.779705Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-06-19T16:10:17.779705Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"0GDnerp5+uVzTXHXEFyxqGn18iIeXaIM61Ro31ssuBdAO4M/znJozeK7YwzLv58aMYX0THPkoIn0vZI8EkLzCw==","signature_status":"signed_v1","signed_at":"2026-06-19T16:10:17.780044Z","signed_message":"canonical_sha256_bytes"},"source_id":"2606.17634","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:d433a1eef87678e8ab57d118ffb6da749acdeaf4b913fbd65101d3632b189508","sha256:2dd695352abc56413ad5ee932b3d26b7aa34368e7592d8fe396af4c9fc0a68d5"],"state_sha256":"6c4f903fcaa6e1bea510372ea58322fb0708cb6202affc13569038dc548a9021"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"XHmC8I5Y6SjNuMpSxo+fjx3Imj0Uj1oi8eamQD01/K/2RXvjUHwAMEBnkRPIWVYKxawfAIxTiABVGzsHLcJAAA==","signed_message":"bundle_sha256_bytes","signed_at":"2026-07-04T15:27:35.017167Z","bundle_sha256":"5afbb5a7006f0e93867287940866a3ebb9dbcb7c282fea0dba9da55d8e41986d"}}