{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:4EFBXV5QKAG6FJKN4MYSUZTYUI","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":"11889dd113b5bf57af7664ad9732b6293265b27677d669440d78a09010c7f1ec","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-09-25T19:40:36Z","title_canon_sha256":"e462750fe38134c39ae25fd21966e1e44481aca2a0389a24997f434b550e7dcb"},"schema_version":"1.0","source":{"id":"2509.21474","kind":"arxiv","version":4}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2509.21474","created_at":"2026-06-02T01:03:33Z"},{"alias_kind":"arxiv_version","alias_value":"2509.21474v4","created_at":"2026-06-02T01:03:33Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2509.21474","created_at":"2026-06-02T01:03:33Z"},{"alias_kind":"pith_short_12","alias_value":"4EFBXV5QKAG6","created_at":"2026-06-02T01:03:33Z"},{"alias_kind":"pith_short_16","alias_value":"4EFBXV5QKAG6FJKN","created_at":"2026-06-02T01:03:33Z"},{"alias_kind":"pith_short_8","alias_value":"4EFBXV5Q","created_at":"2026-06-02T01:03:33Z"}],"graph_snapshots":[{"event_id":"sha256:b5587d0b1c5bc9c272ee6c5a1bc6b4d5620d46e62e1c76800af43b96cb05df3e","target":"graph","created_at":"2026-06-02T01:03:33Z","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/2509.21474/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"While diffusion language models (DLMs) have achieved competitive performance in text generation, improving their reasoning ability with reinforcement learning remains an active research area. Here, we introduce d2, a reasoning framework tailored for masked DLMs. Central to our framework is a new policy gradient algorithm that relies on accurate estimates of the sampling trajectory likelihoods. Because computing these likelihoods naively is computationally expensive for masked DLMs, we develop a family of estimators tailored to distinct model classes. For DLMs that support a sampling algorithm ","authors_text":"Gilad Turok, Guanghan Wang, Marianne Arriola, Volodymyr Kuleshov, Yair Schiff","cross_cats":[],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-09-25T19:40:36Z","title":"d2: Improving Reasoning in Diffusion Language Models via Trajectory Likelihood Estimation"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2509.21474","kind":"arxiv","version":4},"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:baedb7da31905c83f6ab98eecd4c94fdf9e1af25b8115d62a42a8b241a1081b2","target":"record","created_at":"2026-06-02T01:03:33Z","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":"11889dd113b5bf57af7664ad9732b6293265b27677d669440d78a09010c7f1ec","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-09-25T19:40:36Z","title_canon_sha256":"e462750fe38134c39ae25fd21966e1e44481aca2a0389a24997f434b550e7dcb"},"schema_version":"1.0","source":{"id":"2509.21474","kind":"arxiv","version":4}},"canonical_sha256":"e10a1bd7b0500de2a54de3312a6678a22032d124fd31dd4a1c8825d904fbab84","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"e10a1bd7b0500de2a54de3312a6678a22032d124fd31dd4a1c8825d904fbab84","first_computed_at":"2026-06-02T01:03:33.831928Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-06-02T01:03:33.831928Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"oCAui1DsSNPW2+6bF2CQro3VswAg5/PEzuoGB0HWkaW5YtqrDBx2+nuyk8EJUXHjV9U5cOP4ZbUD5dSlBRXJDg==","signature_status":"signed_v1","signed_at":"2026-06-02T01:03:33.832496Z","signed_message":"canonical_sha256_bytes"},"source_id":"2509.21474","source_kind":"arxiv","source_version":4}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:baedb7da31905c83f6ab98eecd4c94fdf9e1af25b8115d62a42a8b241a1081b2","sha256:b5587d0b1c5bc9c272ee6c5a1bc6b4d5620d46e62e1c76800af43b96cb05df3e"],"state_sha256":"56cc879e29f1ab6ad1262d61cd3b9c64cd5aba1d2c49d857769aac766f6d9e27"}