{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:MEZRKB2A6YTLCP4WSBPZZJI7KI","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":"e9f0b9d0cfad68b59b59de90043f57d7c80bb2952bf894ed02bbe9ca961e08e3","cross_cats_sorted":["cs.AI","cs.CL"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-06-30T09:03:24Z","title_canon_sha256":"5d94d20a974a4779b73857a53856bd02f1109f8340cfb77d3f950bb0a506909d"},"schema_version":"1.0","source":{"id":"2606.31371","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2606.31371","created_at":"2026-07-01T01:18:00Z"},{"alias_kind":"arxiv_version","alias_value":"2606.31371v1","created_at":"2026-07-01T01:18:00Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2606.31371","created_at":"2026-07-01T01:18:00Z"},{"alias_kind":"pith_short_12","alias_value":"MEZRKB2A6YTL","created_at":"2026-07-01T01:18:00Z"},{"alias_kind":"pith_short_16","alias_value":"MEZRKB2A6YTLCP4W","created_at":"2026-07-01T01:18:00Z"},{"alias_kind":"pith_short_8","alias_value":"MEZRKB2A","created_at":"2026-07-01T01:18:00Z"}],"graph_snapshots":[{"event_id":"sha256:f87239865cca85a752e63290e5f7a99c7411e12b9dcfc5931be2f258039c70bd","target":"graph","created_at":"2026-07-01T01:18:00Z","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/2606.31371/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"When large language model (LLM) agents adapt their behavior through evaluator feedback, systematic evaluator biases propagate into the agent's learned strategy distribution - a phenomenon termed evaluator preference coupling. Prior work has documented this coupling and established a diagnostic framework (EPC) to measure it, but has not investigated whether calibration techniques can mitigate the effect. We present the first study of evaluator calibration as mitigation: applying probability calibration to the evaluator's pairwise judgments to reduce spurious preference propagation. In a control","authors_text":"Zewen Liu","cross_cats":["cs.AI","cs.CL"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-06-30T09:03:24Z","title":"Calibrating the Evaluator: Does Probability Calibration Mitigate Preference Coupling in LLM Agent Feedback Loops?"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2606.31371","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:b10a2402fbdfc8dcbb16b08268dfdb777de3dbd329f5821871c7395d4a7c85f8","target":"record","created_at":"2026-07-01T01:18:00Z","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":"e9f0b9d0cfad68b59b59de90043f57d7c80bb2952bf894ed02bbe9ca961e08e3","cross_cats_sorted":["cs.AI","cs.CL"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-06-30T09:03:24Z","title_canon_sha256":"5d94d20a974a4779b73857a53856bd02f1109f8340cfb77d3f950bb0a506909d"},"schema_version":"1.0","source":{"id":"2606.31371","kind":"arxiv","version":1}},"canonical_sha256":"6133150740f626b13f96905f9ca51f5225647fba2098e8be8ea2d81bc7375b6b","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"6133150740f626b13f96905f9ca51f5225647fba2098e8be8ea2d81bc7375b6b","first_computed_at":"2026-07-01T01:18:00.991263Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-01T01:18:00.991263Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"C0z16OKvS3FTrcJgb+MbjnRWOeDjtoPYEqr5wiLHFPpimxID2GC0qydliWKQ1nkH1HxiebRX6KEhQ9wqL3j3Cw==","signature_status":"signed_v1","signed_at":"2026-07-01T01:18:00.991717Z","signed_message":"canonical_sha256_bytes"},"source_id":"2606.31371","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:b10a2402fbdfc8dcbb16b08268dfdb777de3dbd329f5821871c7395d4a7c85f8","sha256:f87239865cca85a752e63290e5f7a99c7411e12b9dcfc5931be2f258039c70bd"],"state_sha256":"9273ef5aa0ad80be0aee2af6a7ad6bde8e081aaa1340e22b3c5771314c18bffe"}