{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:YQCAPNL52FXP7VZR6FJ7UNXU5B","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":"60b484058794e4ea407f5ff72619c7b1c3728cfea5986cc205db545da0cade4e","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2026-06-03T17:27:16Z","title_canon_sha256":"d6160d00f9bfca3258d8bd5aef38fcbb5eeca0728a71b4c58ec7616170a21960"},"schema_version":"1.0","source":{"id":"2606.05122","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2606.05122","created_at":"2026-06-04T01:10:07Z"},{"alias_kind":"arxiv_version","alias_value":"2606.05122v1","created_at":"2026-06-04T01:10:07Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2606.05122","created_at":"2026-06-04T01:10:07Z"},{"alias_kind":"pith_short_12","alias_value":"YQCAPNL52FXP","created_at":"2026-06-04T01:10:07Z"},{"alias_kind":"pith_short_16","alias_value":"YQCAPNL52FXP7VZR","created_at":"2026-06-04T01:10:07Z"},{"alias_kind":"pith_short_8","alias_value":"YQCAPNL5","created_at":"2026-06-04T01:10:07Z"}],"graph_snapshots":[{"event_id":"sha256:fa19e0b17e6d270688435bef9f1e0d004829dc5d528439181e8210a51b77cc45","target":"graph","created_at":"2026-06-04T01:10:07Z","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/2606.05122/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Large language models are increasingly evaluated by other models, raising a natural question: can a model predict how a judge will score its own output? We find that the ability is largely present before any targeted training: prompted few-shot, a base model already predicts an external judge's multi-attribute quality scores on open-ended responses well above chance across three benchmarks. We introduce Self-Evaluation Elicitation (SEE), a method that surfaces this latent ability through a short cycle comprising a calibration-coupled reinforcement learning phase that improves the answer and pr","authors_text":"Junfeng Fang, XiuYu Zhang, Yi Shan, Zhenkai Liang","cross_cats":[],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2026-06-03T17:27:16Z","title":"Self-Evaluation Is Already There: Eliciting Latent Judge Calibration in Base LLMs with Minimal Data"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2606.05122","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:ecaf2bba1f562766692f0c5d3fbb5c7afffb4c9d39e1779e3c46e0de931cc0f1","target":"record","created_at":"2026-06-04T01:10:07Z","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":"60b484058794e4ea407f5ff72619c7b1c3728cfea5986cc205db545da0cade4e","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2026-06-03T17:27:16Z","title_canon_sha256":"d6160d00f9bfca3258d8bd5aef38fcbb5eeca0728a71b4c58ec7616170a21960"},"schema_version":"1.0","source":{"id":"2606.05122","kind":"arxiv","version":1}},"canonical_sha256":"c40407b57dd16effd731f153fa36f4e8422d4bfffc1fad4864e7b46201406613","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"c40407b57dd16effd731f153fa36f4e8422d4bfffc1fad4864e7b46201406613","first_computed_at":"2026-06-04T01:10:07.222304Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-06-04T01:10:07.222304Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"CL0b2gafRzQ1NAd7ZUmo8MPledHIsLGucsERhTH0OCCUEM6K/ayvJZZloKGqgKYQORnJeWZWpN45z77y7EmmAg==","signature_status":"signed_v1","signed_at":"2026-06-04T01:10:07.223208Z","signed_message":"canonical_sha256_bytes"},"source_id":"2606.05122","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:ecaf2bba1f562766692f0c5d3fbb5c7afffb4c9d39e1779e3c46e0de931cc0f1","sha256:fa19e0b17e6d270688435bef9f1e0d004829dc5d528439181e8210a51b77cc45"],"state_sha256":"dd96cc0709152deaba3a94c5f8c2483fe7eb40a4c25e658cc0fe7187910cbd27"}