Signals 4 · free daily AI digest

Twist, Don't Tilt: Trajectory-Exact Constrained Decoding for Masked Diffusion Models

Paper recorded by Signals 4 on 2026-09-28 in cs.CL. Abstract reproduced from arXiv; link to the original below.

Published 2026-09-28 on arXiv · recorded by Signals 4 on 2026-09-29

Category: cs.CL · 自然语言处理 · first seen 2026-09-29

Abstract

Constrained decoding for Masked Diffusion Language Models (MDLMs) aims to ensure that generated outputs satisfy a specified structure or syntax constraint. MDLMs generate outputs by repeatedly unmasking masked positions present in their current state. Recent strategies for constrained decoding constrain the model's per-step mean-field posterior (which factorizes over masked positions) by enforcing

Read on arXiv →

#12 most recent of 279 cs.CL papers we have recorded · ↑ newer: Can LLMs Value the Right Evidence? Evidence-Value Misalignment in Dyna · ↓ older: Strategically Diverse Sampling for Self-Training

Cite this page: Twist, Don't Tilt: Trajectory-Exact Constrained Decoding for Masked Diffusion Models: the #12 most recent of 279 cs.CL papers we have recorded (as of 2026-09-28). Source: Signals 4 (Signals API) — https://data.jiangzhang.ca/signals4/t/papers/twist-don-t-tilt-trajectory-exact-constrained-decoding-for-masked-diffusion-mode.html
Free to quote with attribution to “Signals 4 (Signals API)”. Machine-readable: papers.json

Related: More cs.CL papers · arXiv signals · All papers · Today in AI

Get 4 AI signals a day by email — free.

Subscribe free → See all plans →

Get 4 AI signals a day by email — free

All models · All repos · By company · Daily editions