{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:YGCPD5RICGUCZY52RGD6R72RF6","short_pith_number":"pith:YGCPD5RI","schema_version":"1.0","canonical_sha256":"c184f1f62811a82ce3ba8987e8ff512fa218f76a0ed6aca0e6eb314a3234bc3e","source":{"kind":"arxiv","id":"2606.13565","version":1},"attestation_state":"computed","paper":{"title":"A2D2: Fine-Tuning Any-Length Discrete Diffusion for Adaptive Decoding","license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Molei Tao, Pranam Chatterjee, Sophia Tang, Yuchen Zhu","submitted_at":"2026-06-11T16:50:58Z","abstract_excerpt":"Discrete diffusion models offer a simple and stable likelihood-based framework for sequence generation, recently extended to any-length settings via token insertion. Principled reward-guided fine-tuning for any-length discrete diffusion, however, remains largely unexplored. We introduce Fine-Tuning Any-Length Discrete Diffusion for Adaptive Decoding (A2D2), a unified framework for reward-guided fine-tuning of any-length discrete diffusion models via joint optimization of the insertion and unmasking policies together with a quality-based inference schedule. We derive the Radon-Nikodym derivativ"},"verification_status":{"content_addressed":true,"pith_receipt":true,"author_attested":false,"weak_author_claims":0,"strong_author_claims":0,"externally_anchored":false,"storage_verified":false,"citation_signatures":0,"replication_records":0,"graph_snapshot":true,"references_resolved":false,"formal_links_present":false},"canonical_record":{"source":{"id":"2606.13565","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","primary_cat":"cs.LG","submitted_at":"2026-06-11T16:50:58Z","cross_cats_sorted":[],"title_canon_sha256":"0bb3cbf266aa226a63f8c9cda3d96549be2d49784c73686589868844bf6fd3a3","abstract_canon_sha256":"5393d3e055bf9ac532cda510a60f1f714c939503cf9a03042cadfaf4b3ec4bdd"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-12T01:10:10.461497Z","signature_b64":"MvPS0eqa+gv9MYMx435iSEtm+/OhFpFRoWYJ6tE3bVfW+Qme7SizOXtZHf32ECjSE4zoQNCChEmmOjDR6EBVDA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"c184f1f62811a82ce3ba8987e8ff512fa218f76a0ed6aca0e6eb314a3234bc3e","last_reissued_at":"2026-06-12T01:10:10.460585Z","signature_status":"signed_v1","first_computed_at":"2026-06-12T01:10:10.460585Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"A2D2: Fine-Tuning Any-Length Discrete Diffusion for Adaptive Decoding","license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Molei Tao, Pranam Chatterjee, Sophia Tang, Yuchen Zhu","submitted_at":"2026-06-11T16:50:58Z","abstract_excerpt":"Discrete diffusion models offer a simple and stable likelihood-based framework for sequence generation, recently extended to any-length settings via token insertion. Principled reward-guided fine-tuning for any-length discrete diffusion, however, remains largely unexplored. We introduce Fine-Tuning Any-Length Discrete Diffusion for Adaptive Decoding (A2D2), a unified framework for reward-guided fine-tuning of any-length discrete diffusion models via joint optimization of the insertion and unmasking policies together with a quality-based inference schedule. We derive the Radon-Nikodym derivativ"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2606.13565","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/2606.13565/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"},"aliases":[{"alias_kind":"arxiv","alias_value":"2606.13565","created_at":"2026-06-12T01:10:10.460742+00:00"},{"alias_kind":"arxiv_version","alias_value":"2606.13565v1","created_at":"2026-06-12T01:10:10.460742+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2606.13565","created_at":"2026-06-12T01:10:10.460742+00:00"},{"alias_kind":"pith_short_12","alias_value":"YGCPD5RICGUC","created_at":"2026-06-12T01:10:10.460742+00:00"},{"alias_kind":"pith_short_16","alias_value":"YGCPD5RICGUCZY52","created_at":"2026-06-12T01:10:10.460742+00:00"},{"alias_kind":"pith_short_8","alias_value":"YGCPD5RI","created_at":"2026-06-12T01:10:10.460742+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/YGCPD5RICGUCZY52RGD6R72RF6","json":"https://pith.science/pith/YGCPD5RICGUCZY52RGD6R72RF6.json","graph_json":"https://pith.science/api/pith-number/YGCPD5RICGUCZY52RGD6R72RF6/graph.json","events_json":"https://pith.science/api/pith-number/YGCPD5RICGUCZY52RGD6R72RF6/events.json","paper":"https://pith.science/paper/YGCPD5RI"},"agent_actions":{"view_html":"https://pith.science/pith/YGCPD5RICGUCZY52RGD6R72RF6","download_json":"https://pith.science/pith/YGCPD5RICGUCZY52RGD6R72RF6.json","view_paper":"https://pith.science/paper/YGCPD5RI","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2606.13565&json=true","fetch_graph":"https://pith.science/api/pith-number/YGCPD5RICGUCZY52RGD6R72RF6/graph.json","fetch_events":"https://pith.science/api/pith-number/YGCPD5RICGUCZY52RGD6R72RF6/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/YGCPD5RICGUCZY52RGD6R72RF6/action/timestamp_anchor","attest_storage":"https://pith.science/pith/YGCPD5RICGUCZY52RGD6R72RF6/action/storage_attestation","attest_author":"https://pith.science/pith/YGCPD5RICGUCZY52RGD6R72RF6/action/author_attestation","sign_citation":"https://pith.science/pith/YGCPD5RICGUCZY52RGD6R72RF6/action/citation_signature","submit_replication":"https://pith.science/pith/YGCPD5RICGUCZY52RGD6R72RF6/action/replication_record"}},"created_at":"2026-06-12T01:10:10.460742+00:00","updated_at":"2026-06-12T01:10:10.460742+00:00"}