← back
Accepted — Poster
RSS 2026 Workshop on Diffusion Models for Reinforcement Learning (Diff4RL)
When Life Doesn't Give You Q-Values: Diagnosing Q-Exploitation in Test-Time Guidance of Flow-Matching Policies
Prajwal Avhad
<div class="pdf-fallback" style="display:block;"> PDF cannot be displayed in this browser. <br> <a href="/papers/qexploit.pdf">Download PDF</a> </div>