{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:OBGNRHR7WWH7NTHYAOUXIRSJQO","short_pith_number":"pith:OBGNRHR7","schema_version":"1.0","canonical_sha256":"704cd89e3fb58ff6ccf803a974464983b4416eaaff76a6d523aa48536b55c9f3","source":{"kind":"arxiv","id":"2601.11214","version":5},"attestation_state":"computed","paper":{"title":"T$^\\star$: Progressive Block Scaling for Masked Diffusion Language Models Through Trajectory Aware Reinforcement Learning","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Baoyou Chen, Guojiang Zhao, Hanchen Xia, Siyu Zhu, Yutang Ge","submitted_at":"2026-01-16T11:44:12Z","abstract_excerpt":"We present T$^\\star$, a simple TraceRL-based training curriculum for progressive block-size scaling in masked diffusion language models (MDMs). Starting from an AR-initialized small-block MDM, T$^\\star$ transitions smoothly to larger blocks, enabling higher-parallelism decoding with minimal performance degradation on math reasoning benchmarks. Moreover, further analysis suggests that T$^\\star$ may actually converge to an alternative decoding schedule that achieves comparable performance."},"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":"2601.11214","kind":"arxiv","version":5},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2026-01-16T11:44:12Z","cross_cats_sorted":[],"title_canon_sha256":"6311f2181ed8e6267b8260830abebd05da66ba168689cd195104fe394d41bfca","abstract_canon_sha256":"5d83ff0047dfb7797aada8269cc2c97e8883e7bfeaa1057e92f698c80cd7e8fa"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-04T01:08:40.088284Z","signature_b64":"ZxiTSSsiXHw7QVQbHAn6mpj0LY0PQZ1C5z30H+QJMcZojbIiLIWP4Yesb9bAXUQbAAV6TssREsjm04GXP+HcBg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"704cd89e3fb58ff6ccf803a974464983b4416eaaff76a6d523aa48536b55c9f3","last_reissued_at":"2026-06-04T01:08:40.087426Z","signature_status":"signed_v1","first_computed_at":"2026-06-04T01:08:40.087426Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"T$^\\star$: Progressive Block Scaling for Masked Diffusion Language Models Through Trajectory Aware Reinforcement Learning","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Baoyou Chen, Guojiang Zhao, Hanchen Xia, Siyu Zhu, Yutang Ge","submitted_at":"2026-01-16T11:44:12Z","abstract_excerpt":"We present T$^\\star$, a simple TraceRL-based training curriculum for progressive block-size scaling in masked diffusion language models (MDMs). Starting from an AR-initialized small-block MDM, T$^\\star$ transitions smoothly to larger blocks, enabling higher-parallelism decoding with minimal performance degradation on math reasoning benchmarks. Moreover, further analysis suggests that T$^\\star$ may actually converge to an alternative decoding schedule that achieves comparable performance."},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2601.11214","kind":"arxiv","version":5},"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/2601.11214/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":"2601.11214","created_at":"2026-06-04T01:08:40.087544+00:00"},{"alias_kind":"arxiv_version","alias_value":"2601.11214v5","created_at":"2026-06-04T01:08:40.087544+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2601.11214","created_at":"2026-06-04T01:08:40.087544+00:00"},{"alias_kind":"pith_short_12","alias_value":"OBGNRHR7WWH7","created_at":"2026-06-04T01:08:40.087544+00:00"},{"alias_kind":"pith_short_16","alias_value":"OBGNRHR7WWH7NTHY","created_at":"2026-06-04T01:08:40.087544+00:00"},{"alias_kind":"pith_short_8","alias_value":"OBGNRHR7","created_at":"2026-06-04T01:08:40.087544+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2604.16514","citing_title":"BARD: Bridging AutoRegressive and Diffusion Vision-Language Models Via Highly Efficient Progressive Block Merging and Stage-Wise Distillation","ref_index":22,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/OBGNRHR7WWH7NTHYAOUXIRSJQO","json":"https://pith.science/pith/OBGNRHR7WWH7NTHYAOUXIRSJQO.json","graph_json":"https://pith.science/api/pith-number/OBGNRHR7WWH7NTHYAOUXIRSJQO/graph.json","events_json":"https://pith.science/api/pith-number/OBGNRHR7WWH7NTHYAOUXIRSJQO/events.json","paper":"https://pith.science/paper/OBGNRHR7"},"agent_actions":{"view_html":"https://pith.science/pith/OBGNRHR7WWH7NTHYAOUXIRSJQO","download_json":"https://pith.science/pith/OBGNRHR7WWH7NTHYAOUXIRSJQO.json","view_paper":"https://pith.science/paper/OBGNRHR7","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2601.11214&json=true","fetch_graph":"https://pith.science/api/pith-number/OBGNRHR7WWH7NTHYAOUXIRSJQO/graph.json","fetch_events":"https://pith.science/api/pith-number/OBGNRHR7WWH7NTHYAOUXIRSJQO/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/OBGNRHR7WWH7NTHYAOUXIRSJQO/action/timestamp_anchor","attest_storage":"https://pith.science/pith/OBGNRHR7WWH7NTHYAOUXIRSJQO/action/storage_attestation","attest_author":"https://pith.science/pith/OBGNRHR7WWH7NTHYAOUXIRSJQO/action/author_attestation","sign_citation":"https://pith.science/pith/OBGNRHR7WWH7NTHYAOUXIRSJQO/action/citation_signature","submit_replication":"https://pith.science/pith/OBGNRHR7WWH7NTHYAOUXIRSJQO/action/replication_record"}},"created_at":"2026-06-04T01:08:40.087544+00:00","updated_at":"2026-06-04T01:08:40.087544+00:00"}