{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2026:26LCY3DVBRIJ32XQ3A2UQOGRXW","short_pith_number":"pith:26LCY3DV","canonical_record":{"source":{"id":"2606.04396","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2026-06-03T03:22:54Z","cross_cats_sorted":[],"title_canon_sha256":"0f19b842fca88948aa4ccebd01554fe5efefdc7c001980b19ee40f217a1994d3","abstract_canon_sha256":"b04051c6adfa7cd9054f837e5e184a80b36ec28501fa27e06503a64b78726fca"},"schema_version":"1.0"},"canonical_sha256":"d7962c6c750c509deaf0d8354838d1bd91a47c997aa6de20fac607e7719e090a","source":{"kind":"arxiv","id":"2606.04396","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2606.04396","created_at":"2026-06-04T01:09:06Z"},{"alias_kind":"arxiv_version","alias_value":"2606.04396v1","created_at":"2026-06-04T01:09:06Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2606.04396","created_at":"2026-06-04T01:09:06Z"},{"alias_kind":"pith_short_12","alias_value":"26LCY3DVBRIJ","created_at":"2026-06-04T01:09:06Z"},{"alias_kind":"pith_short_16","alias_value":"26LCY3DVBRIJ32XQ","created_at":"2026-06-04T01:09:06Z"},{"alias_kind":"pith_short_8","alias_value":"26LCY3DV","created_at":"2026-06-04T01:09:06Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2026:26LCY3DVBRIJ32XQ3A2UQOGRXW","target":"record","payload":{"canonical_record":{"source":{"id":"2606.04396","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2026-06-03T03:22:54Z","cross_cats_sorted":[],"title_canon_sha256":"0f19b842fca88948aa4ccebd01554fe5efefdc7c001980b19ee40f217a1994d3","abstract_canon_sha256":"b04051c6adfa7cd9054f837e5e184a80b36ec28501fa27e06503a64b78726fca"},"schema_version":"1.0"},"canonical_sha256":"d7962c6c750c509deaf0d8354838d1bd91a47c997aa6de20fac607e7719e090a","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-04T01:09:06.488110Z","signature_b64":"Tbb99gAp6GYbrmXb3RUq4hh/RJa8iJ8Qro0qAi1ou6P9wER3DM/sOoxIUhhupUuJJ99LE9OrhCMtBoL2D1IYBQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"d7962c6c750c509deaf0d8354838d1bd91a47c997aa6de20fac607e7719e090a","last_reissued_at":"2026-06-04T01:09:06.487554Z","signature_status":"signed_v1","first_computed_at":"2026-06-04T01:09:06.487554Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2606.04396","source_version":1,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-06-04T01:09:06Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"Mae/pHMqSLW4Iy3TlSwAG/JUaAU8bpE+lCFCui+rxcFckj5hTJyPkhymx7tWSgv22VXnsag72RWyFEnRbIvnBA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-06-30T07:21:48.062534Z"},"content_sha256":"9e6708c2d3a33c925c8d6aa6280156f9b6de50b9a11c42d7f21c19a06451506a","schema_version":"1.0","event_id":"sha256:9e6708c2d3a33c925c8d6aa6280156f9b6de50b9a11c42d7f21c19a06451506a"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2026:26LCY3DVBRIJ32XQ3A2UQOGRXW","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Read the Trace, Steer the Path: Trajectory-Aware Reinforcement Learning for Diffusion Language Models","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Anant Khandelwal, Manish Gupta","submitted_at":"2026-06-03T03:22:54Z","abstract_excerpt":"Diffusion large language models (dLLMs) generate responses by iteratively unmasking and revising many positions in parallel. This process leaves a rich denoising trace depicting which tokens become confident, which remain unstable, and when commitments form. Existing dLLM reinforcement learning methods use this signal only weakly. Flat rollouts are cheap, but assign a single outcome reward to the whole trajectory. Tree rollouts provide finer, verifiable training signals by branching partial trajectories and propagating leaf rewards upward, but are compute intensive. We ask whether the denoisin"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2606.04396","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/2606.04396/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"},"verdict_id":null},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-06-04T01:09:06Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"kkaxnWcmtg6f/5vMmTkaV92FD2Wf23Xnehpdkr50W2LB3zOFu/5W5QyoWLyRMyAtTjyhrwoHVIgsK/1KBfu5CA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-06-30T07:21:48.062906Z"},"content_sha256":"21171d8b50026df34ad0e94ac3ff62b891af80cbe561cfc4b2ad355041ff7643","schema_version":"1.0","event_id":"sha256:21171d8b50026df34ad0e94ac3ff62b891af80cbe561cfc4b2ad355041ff7643"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/26LCY3DVBRIJ32XQ3A2UQOGRXW/bundle.json","state_url":"https://pith.science/pith/26LCY3DVBRIJ32XQ3A2UQOGRXW/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/26LCY3DVBRIJ32XQ3A2UQOGRXW/bundle.json","status":"primary"}],"public_keys":[{"key_id":"pith-v1-2026-05","algorithm":"ed25519","format":"raw","public_key_b64":"stVStoiQhXFxp4s2pdzPNoqVNBMojDU/fJ2db5S3CbM=","public_key_hex":"b2d552b68890857171a78b36a5dccf368a953413288c353f7c9d9d6f94b709b3","fingerprint_sha256_b32_first128bits":"RVFV5Z2OI2J3ZUO7ERDEBCYNKS","fingerprint_sha256_hex":"8d4b5ee74e4693bcd1df2446408b0d54","rotates_at":null,"url":"https://pith.science/pith-signing-key.json","notes":"Pith uses this Ed25519 key to sign canonical record SHA-256 digests. Verify with: ed25519_verify(public_key, message=canonical_sha256_bytes, signature=base64decode(signature_b64))."}],"merge_version":"pith-open-graph-merge-v1","built_at":"2026-06-30T07:21:48Z","links":{"resolver":"https://pith.science/pith/26LCY3DVBRIJ32XQ3A2UQOGRXW","bundle":"https://pith.science/pith/26LCY3DVBRIJ32XQ3A2UQOGRXW/bundle.json","state":"https://pith.science/pith/26LCY3DVBRIJ32XQ3A2UQOGRXW/state.json","well_known_bundle":"https://pith.science/.well-known/pith/26LCY3DVBRIJ32XQ3A2UQOGRXW/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:26LCY3DVBRIJ32XQ3A2UQOGRXW","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":"b04051c6adfa7cd9054f837e5e184a80b36ec28501fa27e06503a64b78726fca","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2026-06-03T03:22:54Z","title_canon_sha256":"0f19b842fca88948aa4ccebd01554fe5efefdc7c001980b19ee40f217a1994d3"},"schema_version":"1.0","source":{"id":"2606.04396","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2606.04396","created_at":"2026-06-04T01:09:06Z"},{"alias_kind":"arxiv_version","alias_value":"2606.04396v1","created_at":"2026-06-04T01:09:06Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2606.04396","created_at":"2026-06-04T01:09:06Z"},{"alias_kind":"pith_short_12","alias_value":"26LCY3DVBRIJ","created_at":"2026-06-04T01:09:06Z"},{"alias_kind":"pith_short_16","alias_value":"26LCY3DVBRIJ32XQ","created_at":"2026-06-04T01:09:06Z"},{"alias_kind":"pith_short_8","alias_value":"26LCY3DV","created_at":"2026-06-04T01:09:06Z"}],"graph_snapshots":[{"event_id":"sha256:21171d8b50026df34ad0e94ac3ff62b891af80cbe561cfc4b2ad355041ff7643","target":"graph","created_at":"2026-06-04T01:09:06Z","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/2606.04396/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Diffusion large language models (dLLMs) generate responses by iteratively unmasking and revising many positions in parallel. This process leaves a rich denoising trace depicting which tokens become confident, which remain unstable, and when commitments form. Existing dLLM reinforcement learning methods use this signal only weakly. Flat rollouts are cheap, but assign a single outcome reward to the whole trajectory. Tree rollouts provide finer, verifiable training signals by branching partial trajectories and propagating leaf rewards upward, but are compute intensive. We ask whether the denoisin","authors_text":"Anant Khandelwal, Manish Gupta","cross_cats":[],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2026-06-03T03:22:54Z","title":"Read the Trace, Steer the Path: Trajectory-Aware Reinforcement Learning for Diffusion Language Models"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2606.04396","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":"sha256:9e6708c2d3a33c925c8d6aa6280156f9b6de50b9a11c42d7f21c19a06451506a","target":"record","created_at":"2026-06-04T01:09:06Z","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":"b04051c6adfa7cd9054f837e5e184a80b36ec28501fa27e06503a64b78726fca","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2026-06-03T03:22:54Z","title_canon_sha256":"0f19b842fca88948aa4ccebd01554fe5efefdc7c001980b19ee40f217a1994d3"},"schema_version":"1.0","source":{"id":"2606.04396","kind":"arxiv","version":1}},"canonical_sha256":"d7962c6c750c509deaf0d8354838d1bd91a47c997aa6de20fac607e7719e090a","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"d7962c6c750c509deaf0d8354838d1bd91a47c997aa6de20fac607e7719e090a","first_computed_at":"2026-06-04T01:09:06.487554Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-06-04T01:09:06.487554Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"Tbb99gAp6GYbrmXb3RUq4hh/RJa8iJ8Qro0qAi1ou6P9wER3DM/sOoxIUhhupUuJJ99LE9OrhCMtBoL2D1IYBQ==","signature_status":"signed_v1","signed_at":"2026-06-04T01:09:06.488110Z","signed_message":"canonical_sha256_bytes"},"source_id":"2606.04396","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:9e6708c2d3a33c925c8d6aa6280156f9b6de50b9a11c42d7f21c19a06451506a","sha256:21171d8b50026df34ad0e94ac3ff62b891af80cbe561cfc4b2ad355041ff7643"],"state_sha256":"f220f39fe8f8f37afc311555317ba831b86bf611f26640971722ee97b80af813"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"rU0cAvzsI1M0bfzY25IGG+YZu2t8FE41GBio0gXVpH97BXQeIa51SV5ZLK0dGkw4xCf098LNSVFaMfJCpo2QDg==","signed_message":"bundle_sha256_bytes","signed_at":"2026-06-30T07:21:48.064949Z","bundle_sha256":"fdf57e8666282466da1cd83e713c5a0ce22e2353d065d78d99287d2fff59ccea"}}