{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:26OFLDTLSEV7LRK74ZLXOKQBGI","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":"ffbf0ffeaefd483e7f45bb3473c1f7e1473c27f7419bfc15da877e91324039de","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-05-28T05:47:40Z","title_canon_sha256":"dde5a7c57de24e903572613750bf63f443ac80ee37e380f7b6474f992581ad8e"},"schema_version":"1.0","source":{"id":"2605.29398","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2605.29398","created_at":"2026-05-29T01:05:37Z"},{"alias_kind":"arxiv_version","alias_value":"2605.29398v1","created_at":"2026-05-29T01:05:37Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.29398","created_at":"2026-05-29T01:05:37Z"},{"alias_kind":"pith_short_12","alias_value":"26OFLDTLSEV7","created_at":"2026-05-29T01:05:37Z"},{"alias_kind":"pith_short_16","alias_value":"26OFLDTLSEV7LRK7","created_at":"2026-05-29T01:05:37Z"},{"alias_kind":"pith_short_8","alias_value":"26OFLDTL","created_at":"2026-05-29T01:05:37Z"}],"graph_snapshots":[{"event_id":"sha256:9ed9fb2c64e636efec74c10781b075ee64b65fd25e83fb98d347c93dd12b41da","target":"graph","created_at":"2026-05-29T01:05:37Z","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/2605.29398/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Reinforcement learning (RL) can be used to improve the policy (denoiser) of diffusion large language models (dLLMs), while being hindered by the intractability of the policy likelihood. A dominant and efficient family of methods replaces the likelihood in standard RL with its evidence lower bound (ELBO), estimated from randomly masked sequences. Despite being well aligned with pre-training, these approaches introduce bias through training--inference mismatch by using the ELBO as a likelihood surrogate, which can degrade performance. In this work, we propose Guided Denoiser Self-Distillation (G","authors_text":"Che Liu, Ilija Bogunovic, Keyue Jiang, Qifang Zhao, Sangwoong Yoon, Xiaohang Tang, Xiaoxiao Xu","cross_cats":["cs.AI"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-05-28T05:47:40Z","title":"GDSD: Reinforcement Learning as Guided Denoiser Self-Distillation for Diffusion Language Models"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2605.29398","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":"sha256:4ef5f66b7b349ce3a94385b135b31a2ff49cde70008f16e193a4c090f4fd39ed","target":"record","created_at":"2026-05-29T01:05:37Z","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":"ffbf0ffeaefd483e7f45bb3473c1f7e1473c27f7419bfc15da877e91324039de","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-05-28T05:47:40Z","title_canon_sha256":"dde5a7c57de24e903572613750bf63f443ac80ee37e380f7b6474f992581ad8e"},"schema_version":"1.0","source":{"id":"2605.29398","kind":"arxiv","version":1}},"canonical_sha256":"d79c558e6b912bf5c55fe657772a01321ca17288cd8a07ece8817c6b5e2850a7","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"d79c558e6b912bf5c55fe657772a01321ca17288cd8a07ece8817c6b5e2850a7","first_computed_at":"2026-05-29T01:05:37.021404Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-29T01:05:37.021404Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"L9kbkS6nZYpXHVLu5f6trxmWHYxt06PKvGQX5O0G9az8N2BaXwRWAifjtutv79CtbK/NPyZ6mCQrNi6RgNiOAw==","signature_status":"signed_v1","signed_at":"2026-05-29T01:05:37.022187Z","signed_message":"canonical_sha256_bytes"},"source_id":"2605.29398","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:4ef5f66b7b349ce3a94385b135b31a2ff49cde70008f16e193a4c090f4fd39ed","sha256:9ed9fb2c64e636efec74c10781b075ee64b65fd25e83fb98d347c93dd12b41da"],"state_sha256":"09043678d08119a0f708683cfcf24b4c791ca22ce94d783fd1d3a98daed7025c"}