Signals 4 · free daily AI digest

Discrete Beckmann Transport Models for One-Step Language Modeling and Reasoning

Paper recorded by Signals 4 on 2026-09-14 in cs.LG. Abstract reproduced from arXiv; link to the original below.

Published 2026-09-14 on arXiv · recorded by Signals 4 on 2026-09-15

Category: cs.LG · 机器学习 · first seen 2026-09-15

Abstract

Discrete diffusion and flow models are a promising alternative to autoregressive language models, but compressing many-step sampling into fewer steps typically requires distilling a pretrained teacher model. This caps the student at the teacher's quality and requires a costly two-stage training pipeline. We introduce Discrete Beckmann Transport Models (DBTM), built on a time-independent flow whose

Read on arXiv →

#50 most recent of 215 cs.LG papers we have recorded · ↑ newer: Safe Meta-Reinforcement Learning via Information Space Reachability · ↓ older: Bridging Control, Inference, Transport, and Thermodynamics: From Theor
Cite this page: Discrete Beckmann Transport Models for One-Step Language Modeling and Reasoning: the #50 most recent of 215 cs.LG papers we have recorded (as of 2026-09-14). Source: Signals 4 (Signals API) — https://data.jiangzhang.ca/signals4/t/papers/discrete-beckmann-transport-models-for-one-step-language-modeling-and-reasoning.html
Free to quote with attribution to “Signals 4 (Signals API)”. Machine-readable: papers.json
Related: More cs.LG 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