{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2026:NFR7D3TKF5FJXHBHE35KARRZL3","short_pith_number":"pith:NFR7D3TK","canonical_record":{"source":{"id":"2605.11854","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2026-05-12T09:39:06Z","cross_cats_sorted":[],"title_canon_sha256":"81b1fcd3984a8bed55856acf3b3d8e6dbe00b7b2cfc8a57fc276301cb0e49e40","abstract_canon_sha256":"dc78199a10256301f0dcd74b28737650a982dce09dcd11f273d65eb6aabd7b7c"},"schema_version":"1.0"},"canonical_sha256":"6963f1ee6a2f4a9b9c2726faa046395efde5b325c11605e8d6fbf224148341e0","source":{"kind":"arxiv","id":"2605.11854","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2605.11854","created_at":"2026-05-20T00:04:36Z"},{"alias_kind":"arxiv_version","alias_value":"2605.11854v2","created_at":"2026-05-20T00:04:36Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.11854","created_at":"2026-05-20T00:04:36Z"},{"alias_kind":"pith_short_12","alias_value":"NFR7D3TKF5FJ","created_at":"2026-05-20T00:04:36Z"},{"alias_kind":"pith_short_16","alias_value":"NFR7D3TKF5FJXHBH","created_at":"2026-05-20T00:04:36Z"},{"alias_kind":"pith_short_8","alias_value":"NFR7D3TK","created_at":"2026-05-20T00:04:36Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2026:NFR7D3TKF5FJXHBHE35KARRZL3","target":"record","payload":{"canonical_record":{"source":{"id":"2605.11854","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2026-05-12T09:39:06Z","cross_cats_sorted":[],"title_canon_sha256":"81b1fcd3984a8bed55856acf3b3d8e6dbe00b7b2cfc8a57fc276301cb0e49e40","abstract_canon_sha256":"dc78199a10256301f0dcd74b28737650a982dce09dcd11f273d65eb6aabd7b7c"},"schema_version":"1.0"},"canonical_sha256":"6963f1ee6a2f4a9b9c2726faa046395efde5b325c11605e8d6fbf224148341e0","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-20T00:04:36.308337Z","signature_b64":"pbGIS9FF3Fbu+muez+vooVkk+EYCAhRYdKOiSK4DecDiIfXLsNU/9lDuESLGT8hirx+YJSztR5y1UdbTmhclCA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"6963f1ee6a2f4a9b9c2726faa046395efde5b325c11605e8d6fbf224148341e0","last_reissued_at":"2026-05-20T00:04:36.307179Z","signature_status":"signed_v1","first_computed_at":"2026-05-20T00:04:36.307179Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2605.11854","source_version":2,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-05-20T00:04:36Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"bzoPA0FgktQrpoZMRUIM+9fnd3CRznZvAbf37WljjLFKv7AOJDtnHJO4dhfftJfpwPMbFiTd066UpqWZacRPAA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-23T16:07:51.087488Z"},"content_sha256":"8513f6d64ce3bbb130479a9b08062e013cf75e8124984920448cf3ea44b36c04","schema_version":"1.0","event_id":"sha256:8513f6d64ce3bbb130479a9b08062e013cf75e8124984920448cf3ea44b36c04"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2026:NFR7D3TKF5FJXHBHE35KARRZL3","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Self-Distilled Trajectory-Aware Boltzmann Modeling: Bridging the Training-Inference Discrepancy in Diffusion Language Models","license":"http://creativecommons.org/licenses/by/4.0/","headline":"A new optimization framework for diffusion language models uses self-distilled inference trajectories and Boltzmann modeling of entropies to close the gap with standard supervised fine-tuning.","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Dandan Tu, Haoliang Li, Hui Liu, Kecheng Chen, Lingpeng Kong, Rui Liu, Shi Wu, Suiyun Zhang, Xijia Tao, Xinyu Fu, Yibing Liu, Ziru Liu","submitted_at":"2026-05-12T09:39:06Z","abstract_excerpt":"Diffusion Language Models (DLMs) have recently emerged as a promising alternative to autoregressive language models, offering stronger global awareness and highly parallel generation. However, post-training DLMs with standard Negative Evidence Lower Bound (NELBO)-based supervised fine-tuning remains inefficient: training reconstructs randomly masked tokens in a single step, whereas inference follows a confidence-guided, multi-step easy-to-hard denoising trajectory. Recent trajectory-based self-distillation methods exploit such inference trajectories mainly for sampling-step compression and acc"},"claims":{"count":4,"items":[{"kind":"strongest_claim","text":"TABOM achieves substantial gains in new domains, expands the effective knowledge boundary of DLMs, and significantly mitigates catastrophic forgetting compared with standard SFT.","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","text":"That modeling the inference unmasking preference as a Boltzmann distribution over predictive entropies and deriving a pairwise ranking objective from it will produce genuine knowledge acquisition rather than marginal or illusory gains.","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","text":"TABOM models inference unmasking preferences as a Boltzmann distribution over predictive entropies and derives a ranking loss to align DLM training with observed trajectories, yielding gains in new domains and reduced catastrophic forgetting versus standard SFT.","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"},{"kind":"headline","text":"A new optimization framework for diffusion language models uses self-distilled inference trajectories and Boltzmann modeling of entropies to close the gap with standard supervised fine-tuning.","source":"verdict.pith_extraction.headline","status":"machine_extracted","claim_id":"C4","attestation":"unclaimed"}],"snapshot_sha256":"39504c315db187e1d938a3635f6991d019c81cf91b39a2672696461f8066250d"},"source":{"id":"2605.11854","kind":"arxiv","version":2},"verdict":{"id":"d6c29f0f-8efb-4c95-87c2-0163a8e682bc","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-13T05:48:46.216040Z","strongest_claim":"TABOM achieves substantial gains in new domains, expands the effective knowledge boundary of DLMs, and significantly mitigates catastrophic forgetting compared with standard SFT.","one_line_summary":"TABOM models inference unmasking preferences as a Boltzmann distribution over predictive entropies and derives a ranking loss to align DLM training with observed trajectories, yielding gains in new domains and reduced catastrophic forgetting versus standard SFT.","pipeline_version":"pith-pipeline@v0.9.0","weakest_assumption":"That modeling the inference unmasking preference as a Boltzmann distribution over predictive entropies and deriving a pairwise ranking objective from it will produce genuine knowledge acquisition rather than marginal or illusory gains.","pith_extraction_headline":"A new optimization framework for diffusion language models uses self-distilled inference trajectories and Boltzmann modeling of entropies to close the gap with standard supervised fine-tuning."},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2605.11854/integrity.json","findings":[],"available":true,"detectors_run":[{"name":"ai_meta_artifact","ran_at":"2026-05-19T11:36:39.823964Z","status":"completed","version":"1.0.0","findings_count":0},{"name":"doi_title_agreement","ran_at":"2026-05-19T09:01:18.338110Z","status":"completed","version":"1.0.0","findings_count":0},{"name":"doi_compliance","ran_at":"2026-05-19T08:08:14.829608Z","status":"completed","version":"1.0.0","findings_count":0}],"snapshot_sha256":"3e31ee0a46068f7e82e0ecf3caef4096db9c92f52d8a0fa93e0f2efbbc1ea3ad"},"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":"d6c29f0f-8efb-4c95-87c2-0163a8e682bc"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-05-20T00:04:36Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"zY33nhbDvIBLRR8JNuRLeAnO01UyLTTeZZ+6p7shfeqoPDLhYAFT1F+EKci5XdznsKMYFHeM05vqr4joOGW5DA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-23T16:07:51.088461Z"},"content_sha256":"ff61761579258bc77662f7becc903f85c26f8ce6cc7405a85e25a68fd349a533","schema_version":"1.0","event_id":"sha256:ff61761579258bc77662f7becc903f85c26f8ce6cc7405a85e25a68fd349a533"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/NFR7D3TKF5FJXHBHE35KARRZL3/bundle.json","state_url":"https://pith.science/pith/NFR7D3TKF5FJXHBHE35KARRZL3/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/NFR7D3TKF5FJXHBHE35KARRZL3/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":"2026-05-23T16:07:51Z","links":{"resolver":"https://pith.science/pith/NFR7D3TKF5FJXHBHE35KARRZL3","bundle":"https://pith.science/pith/NFR7D3TKF5FJXHBHE35KARRZL3/bundle.json","state":"https://pith.science/pith/NFR7D3TKF5FJXHBHE35KARRZL3/state.json","well_known_bundle":"https://pith.science/.well-known/pith/NFR7D3TKF5FJXHBHE35KARRZL3/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:NFR7D3TKF5FJXHBHE35KARRZL3","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":"dc78199a10256301f0dcd74b28737650a982dce09dcd11f273d65eb6aabd7b7c","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2026-05-12T09:39:06Z","title_canon_sha256":"81b1fcd3984a8bed55856acf3b3d8e6dbe00b7b2cfc8a57fc276301cb0e49e40"},"schema_version":"1.0","source":{"id":"2605.11854","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2605.11854","created_at":"2026-05-20T00:04:36Z"},{"alias_kind":"arxiv_version","alias_value":"2605.11854v2","created_at":"2026-05-20T00:04:36Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.11854","created_at":"2026-05-20T00:04:36Z"},{"alias_kind":"pith_short_12","alias_value":"NFR7D3TKF5FJ","created_at":"2026-05-20T00:04:36Z"},{"alias_kind":"pith_short_16","alias_value":"NFR7D3TKF5FJXHBH","created_at":"2026-05-20T00:04:36Z"},{"alias_kind":"pith_short_8","alias_value":"NFR7D3TK","created_at":"2026-05-20T00:04:36Z"}],"graph_snapshots":[{"event_id":"sha256:ff61761579258bc77662f7becc903f85c26f8ce6cc7405a85e25a68fd349a533","target":"graph","created_at":"2026-05-20T00:04:36Z","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":4,"items":[{"attestation":"unclaimed","claim_id":"C1","kind":"strongest_claim","source":"verdict.strongest_claim","status":"machine_extracted","text":"TABOM achieves substantial gains in new domains, expands the effective knowledge boundary of DLMs, and significantly mitigates catastrophic forgetting compared with standard SFT."},{"attestation":"unclaimed","claim_id":"C2","kind":"weakest_assumption","source":"verdict.weakest_assumption","status":"machine_extracted","text":"That modeling the inference unmasking preference as a Boltzmann distribution over predictive entropies and deriving a pairwise ranking objective from it will produce genuine knowledge acquisition rather than marginal or illusory gains."},{"attestation":"unclaimed","claim_id":"C3","kind":"one_line_summary","source":"verdict.one_line_summary","status":"machine_extracted","text":"TABOM models inference unmasking preferences as a Boltzmann distribution over predictive entropies and derives a ranking loss to align DLM training with observed trajectories, yielding gains in new domains and reduced catastrophic forgetting versus standard SFT."},{"attestation":"unclaimed","claim_id":"C4","kind":"headline","source":"verdict.pith_extraction.headline","status":"machine_extracted","text":"A new optimization framework for diffusion language models uses self-distilled inference trajectories and Boltzmann modeling of entropies to close the gap with standard supervised fine-tuning."}],"snapshot_sha256":"39504c315db187e1d938a3635f6991d019c81cf91b39a2672696461f8066250d"},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"integrity":{"available":true,"clean":true,"detectors_run":[{"findings_count":0,"name":"ai_meta_artifact","ran_at":"2026-05-19T11:36:39.823964Z","status":"completed","version":"1.0.0"},{"findings_count":0,"name":"doi_title_agreement","ran_at":"2026-05-19T09:01:18.338110Z","status":"completed","version":"1.0.0"},{"findings_count":0,"name":"doi_compliance","ran_at":"2026-05-19T08:08:14.829608Z","status":"completed","version":"1.0.0"}],"endpoint":"/pith/2605.11854/integrity.json","findings":[],"snapshot_sha256":"3e31ee0a46068f7e82e0ecf3caef4096db9c92f52d8a0fa93e0f2efbbc1ea3ad","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Diffusion Language Models (DLMs) have recently emerged as a promising alternative to autoregressive language models, offering stronger global awareness and highly parallel generation. However, post-training DLMs with standard Negative Evidence Lower Bound (NELBO)-based supervised fine-tuning remains inefficient: training reconstructs randomly masked tokens in a single step, whereas inference follows a confidence-guided, multi-step easy-to-hard denoising trajectory. Recent trajectory-based self-distillation methods exploit such inference trajectories mainly for sampling-step compression and acc","authors_text":"Dandan Tu, Haoliang Li, Hui Liu, Kecheng Chen, Lingpeng Kong, Rui Liu, Shi Wu, Suiyun Zhang, Xijia Tao, Xinyu Fu, Yibing Liu, Ziru Liu","cross_cats":[],"headline":"A new optimization framework for diffusion language models uses self-distilled inference trajectories and Boltzmann modeling of entropies to close the gap with standard supervised fine-tuning.","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2026-05-12T09:39:06Z","title":"Self-Distilled Trajectory-Aware Boltzmann Modeling: Bridging the Training-Inference Discrepancy in Diffusion Language Models"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2605.11854","kind":"arxiv","version":2},"verdict":{"created_at":"2026-05-13T05:48:46.216040Z","id":"d6c29f0f-8efb-4c95-87c2-0163a8e682bc","model_set":{"reader":"grok-4.3"},"one_line_summary":"TABOM models inference unmasking preferences as a Boltzmann distribution over predictive entropies and derives a ranking loss to align DLM training with observed trajectories, yielding gains in new domains and reduced catastrophic forgetting versus standard SFT.","pipeline_version":"pith-pipeline@v0.9.0","pith_extraction_headline":"A new optimization framework for diffusion language models uses self-distilled inference trajectories and Boltzmann modeling of entropies to close the gap with standard supervised fine-tuning.","strongest_claim":"TABOM achieves substantial gains in new domains, expands the effective knowledge boundary of DLMs, and significantly mitigates catastrophic forgetting compared with standard SFT.","weakest_assumption":"That modeling the inference unmasking preference as a Boltzmann distribution over predictive entropies and deriving a pairwise ranking objective from it will produce genuine knowledge acquisition rather than marginal or illusory gains."}},"verdict_id":"d6c29f0f-8efb-4c95-87c2-0163a8e682bc"}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:8513f6d64ce3bbb130479a9b08062e013cf75e8124984920448cf3ea44b36c04","target":"record","created_at":"2026-05-20T00:04:36Z","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":"dc78199a10256301f0dcd74b28737650a982dce09dcd11f273d65eb6aabd7b7c","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2026-05-12T09:39:06Z","title_canon_sha256":"81b1fcd3984a8bed55856acf3b3d8e6dbe00b7b2cfc8a57fc276301cb0e49e40"},"schema_version":"1.0","source":{"id":"2605.11854","kind":"arxiv","version":2}},"canonical_sha256":"6963f1ee6a2f4a9b9c2726faa046395efde5b325c11605e8d6fbf224148341e0","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"6963f1ee6a2f4a9b9c2726faa046395efde5b325c11605e8d6fbf224148341e0","first_computed_at":"2026-05-20T00:04:36.307179Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-20T00:04:36.307179Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"pbGIS9FF3Fbu+muez+vooVkk+EYCAhRYdKOiSK4DecDiIfXLsNU/9lDuESLGT8hirx+YJSztR5y1UdbTmhclCA==","signature_status":"signed_v1","signed_at":"2026-05-20T00:04:36.308337Z","signed_message":"canonical_sha256_bytes"},"source_id":"2605.11854","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:8513f6d64ce3bbb130479a9b08062e013cf75e8124984920448cf3ea44b36c04","sha256:ff61761579258bc77662f7becc903f85c26f8ce6cc7405a85e25a68fd349a533"],"state_sha256":"cdfd07edd1322b6be6f2767bfdc5d332d890d4ca56c386f1fce921ff093f60f5"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"GmjDyPgwM+ZlpQ2AMMWCeVALOx17dP7RlsXpr3n+uuw02QDzt520YEk+25vLRtck4mm5rNV8MFJE201y2oR2DA==","signed_message":"bundle_sha256_bytes","signed_at":"2026-05-23T16:07:51.093194Z","bundle_sha256":"e5de54874e94f057f48b436b3ccf4def6f194645add72e57ab4bef5381aa432f"}}