# Apple 提出 TS-DFM：用能量罗盘引导少步离散流匹配蒸馏

- 来源：Apple Machine Learning Research（RSS）
- 发布时间：2026-09-16 08:00
- AIHOT 分数：45
- AIHOT 链接：https://aihot.news/items/cmu4nc77e06tqrodcauixijp6
- 原文链接：https://machinelearning.apple.com/research/trajectory-teacher-flow-matching

## AI 摘要

Apple 研究团队提出 Trajectory-Shaped Discrete Flow Matching（TS-DFM），用轻量级能量罗盘在训练中评估并筛选中间步骤，替代离散流匹配的盲目随机跳跃，且不增加推理成本。在 170M 参数语言模型上，8 步学生模型困惑度比 1024 步教师模型低 32%，速度快 128×。

## 正文

AuthorsAmin Karimi Monsefi†**, Dominic Culver, Nikhil Bhendawade, Manuel R. Ciosici, Yizhe Zhang, Irina Belousova

Discrete flow matching generates text by iteratively transforming noise tokens into coherent language, but may require hundreds of forward passes. Distillation uses the multi-step trajectory to train a student to reproduce the process in a few steps. When the student underperforms, the usual explanation is insufficient capacity. We argue the opposite: the trajectory is the bottleneck, not the student. Each training trajectory is built through a chain of blind stochastic jumps with no evaluation of sequence quality; a single bad decision at an early midpoint propagates through subsequent steps, yet the student must imitate the result.

Trajectory-Shaped Discrete Flow Matching (TS-DFM) replaces these blind jumps with guided navigation: a lightweight energy compass evaluates candidate continuations at each midpoint, selecting the most coherent. All shaping is training-only; inference cost is unchanged. On 170M-parameter language modeling, the shaped student at 8 steps achieves 32% lower perplexity than the 1,024-step teacher while being 128× faster, with gains consistent across source distributions and three evaluators of increasing scale. TS-DFM achieves the best perplexity of any discrete-generation baseline we compare against, including methods trained on 6× more data or using 5× larger models.

† The Ohio State University

** Work done while at Apple