SetCommit: Coordinated Token Commitments for Few-Step Diffusion Language Model Decoding
Its authors withdrew this paper, so there is no full text to read or summarize.
Lists
Only you see this
Abstract
Diffusion language models generate multiple token positions in parallel, but few-step decoding makes parallel commitment difficult: tokens that appear reliable individually may be jointly inconsistent, while premature commitments can prevent later context-dependent refinement. In this paper, we introduce \textsc{SetCommit}, a training-free method for coordinated parallel token commitment. Instead of selecting tokens independently by confidence, \textsc{SetCommit} selects a commitment set that balances token reliability against local joint commitment risk, preserving conditioning opportunities for subsequent denoising steps. It further performs budget-neutral local exchanges, replacing vulnerable earlier commitments when newly revealed context supports more reliable alternatives. \textsc{SetCommit} requires no retraining or architectural modification and is compatible with existing diffusion decoding strategies. Experiments with LLaDA-8B and Dream-7B on GSM8K, MATH-500, MBPP, and HumanEval show accuracy/pass@1 gains of up to 7.4 percentage points over native decoding at 32 model evaluations. At common quality targets, geometric-mean incremental speedups range from $1.44$ to $1.74\times$. Ablations support the benefits of coordinated selection and bounded remasking. Code is available at \url{https://anonymous.4open.science/r/review-artifact-48dc4648-8B86}.
BibTeX
@misc{feng2027setcommit,
author = {Yuannuo Feng and Yubing Ye and Wenshuai Yao and Yizhe Chen and Ngai Wong and Wenyong Zhou and Wang Kang},
title = {{SetCommit: Coordinated Token Commitments for Few-Step Diffusion Language Model Decoding}},
year = {2027},
url = {https://openreview.net/forum?id=LxhKURNgpg},
howpublished = {OpenReview},
note = {Submitted to ICLR 2027},
}Loading related content...