T$^\star$: Progressive Block Scaling for Masked Diffusion Language Models Through Trajectory Aware Reinforcement Learning

Baoyou Chen; Guojiang Zhao; Hanchen Xia; Siyu Zhu; Yutang Ge

T^star: Progressive Block Scaling for Masked Diffusion Language Models Through Trajectory Aware Reinforcement Learning

Not yet reviewed by Pith; the record is open.

Re-run · record.json Download PDF Read on arXiv ↗

This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.

SPECIMEN: schema-true, not a live event

T0 review · schema-true

One-sentence machine reading of the paper's core claim.

pith:XXXXXXXX · record.json · timestamp

arxiv 2601.11214 v5 pith:OBGNRHR7 submitted 2026-01-16 cs.CL

T^star: Progressive Block Scaling for Masked Diffusion Language Models Through Trajectory Aware Reinforcement Learning

Hanchen Xia , Baoyou Chen , Yutang Ge , Guojiang Zhao , Siyu Zhu This is my paper

classification cs.CL

keywords stardecodingdiffusionlanguagemaskedmodelsperformanceprogressive

verification ladder T0 review T1 audit T2 compute T3 formal T4 reserved

0 comments

read the original abstract

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.

discussion (0)

Forward citations

Cited by 2 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score.

BARD: Bridging AutoRegressive and Diffusion Vision-Language Models Via Highly Efficient Progressive Block Merging and Stage-Wise Distillation
cs.CV 2026-04 unverdicted novelty 7.0

BARD bridges autoregressive and diffusion VLMs with progressive block merging plus stage-wise intra-diffusion distillation, delivering 3x speedup and new SOTA on open dVLMs using under 4.4M data points.
SemBlock: Semantic Boundary Dynamic Blocks for Diffusion LLMs
cs.CL 2026-06 unverdicted novelty 6.0

SemBlock adds semantic-boundary prediction to enable dynamic block decoding in diffusion LLMs and reports gains over fixed-block and AdaBlock baselines on GSM8K, IFEval, MATH, and HumanEval.