Skip to content

Timestep Weighting: A Hidden Key to Effective ELBO-Based Flow-Matching RL

Sep 2026 · 0 citations
Computer Science

Abstract

ELBO-based reinforcement learning offers a sampler-agnostic approach to fine-tuning flow matching models with reward feedback. Timestep weighting in ELBO-based RL has large impact on performance, and it also provides a unified view (as we show in this work) to understand prediction losses heuristically chosen in prior work, yet it remains under-researched and is often chosen to inherit pretrain configs. We investigate impacts and dynamics of timestep weighting in ELBO-based RL. We show that effective weighting depends on both the reward landscape and stage of learning. (1) Through experiments on controlled CIFAR image generation, complemented by robotics, we investigate how weighting impacts reward-driven updates across noise levels. (2) Through gradient analysis, we reveal distinct patterns of cross-noise coordination across tasks and their evolution during training. These findings motivate the hypothesis that useful weighting depends on the gap between the policy's current behavior and the behavior favored by the reward. (3) Guided by this analysis, we study simple static weighting, budgeted profile selection, and dynamic schedules that improve performance beyond conventional target choices. Our results establish timestep weighting as an important design choice for flow-matching RL and motivate further research into methods that choose and adapt it throughout learning.

View source

Similar papers

#artificial intelligence Preprint Sep 2026

MaD-RL: Matching Distributions for Calibrating LLMs with Reinforcement Learning

This work proposes a general RL-based framework for Distribution Matching allowing matching the distribution of a latent categorical attribute of model outputs to a specified target distribution and proposes reward functions for other divergences such as KL and Jensen-Shannon and motivate them with theoretical justific...

Sourabh Kulkarni, Ksheeraj Sai Vepuri, B. Demir et al. · 1 citation
Conference Aug 2026

Post-Training for Reasoning LLMs with Reinforcement Learning: A Stability–Efficiency Perspective

Reinforcement Learning (RL)-based post-training has become a key mechanism for improving reasoning Large Language Models (LLMs), especially when automated verifiers enable Reinforcement Learning with Verifiable Reward (RLVR). Recent methods, however, are increasingly modular: many apparent algorithmic advances recombin...

Liu Yang, Han Zhu, Zheng-Yang Zhong et al. · 2 citations
#machine learning Preprint Oct 2026

Reinforcement Learning for Hierarchical Reasoning Rewards: Minimax-Optimal Rates with Transformers

Reinforcement learning (RL) has become a standard tool for post-training language models on reasoning tasks, where the policy is updated by reward feedback while exploring the space of responses. Despite its empirical success, theoretical understanding of RL post-training remains limited, in particular of why on-policy...

Naoki Nishikawa, Taiji Suzuki · 0 citations

Learning Beyond What You Sample: Off-Policy-Aware Cross-Model Trajectory Exchange for RLVR

GRAFT (Gated Replacement of Answer-Failed groups with peer Trajectories), an off-policy-aware framework that replaces all-fail groups with informative peer groups and transfers both successful and unsuccessful peer responses with peer-computed advantages, while controlling cross-model mismatch through sequence-level co...

Doohyuk Jang, Y. Park, Gyouk Chu et al. · 0 citations
#artificial intelligence Preprint Oct 2026

RLDISCOVER: LLM-driven co-evolution of reinforcement learning algorithms

LLM-guided program evolution has enabled discoveries in mathematics and computational optimization, raising the prospect of reinforcement learning (RL) algorithms that self-evolve to improve how agents learn. However, realizing this prospect faces two obstacles. Joint search over coupled algorithmic components is diffi...

Hao-Ran Li, Z. Ge, Xiao-Ming Yuan et al. · 0 citations
#machine learning Preprint Sep 2026

Fine-Tuning on Self-Generated and Reward-Weighted Data: Learning Dynamics, Convergence Rates, and Benefits of Off-Policyness

A unified theory for RE(S) is developed that covers the full spectrum of S, and can be interpreted as a stage-wise optimization process, where each stage takes $S$ gradient steps for minimizing the Kullback-Leibler distance to a fixed reward-weighted rollout distribution.

Zhi-Wei Wang, Yan-Xi Chen, Ya-Liang Li et al. · 0 citations

Related blog posts

MIT News · Artificial Intelligence Oct 28, 2026

Making machines that can listen

Professor Paris Smaragdis develops computers with superhuman hearing for applications like speech recognition, music understanding, and environmental monitoring.

We use cookies to run the site and, with your consent, for analytics and to show ads. See our Cookie Policy.