paper-with-me

홈 › Papers

Reward Under Attack: Analyzing the Robustness and Hackability of Process Reward Models

2026-02-20 · Rishabh Tiwari, Aditya Tomar, Udbhav Bamba, Monishwaran Maheswaran, Heng Yang, Michael W. Mahoney, Kurt Keutzer, Amir Gholami arxiv

Process Reward Models (PRMs) are rapidly becoming the backbone of LLM reasoning pipelines, yet we demonstrate that state-of-the-art PRMs are systematically exploitable under adversarial optimization pressure. To address this, we introduce a three-tiered diagnostic framework that applies increasing adversarial pressure to quantify these vulnerabilities. Static perturbation analysis uncovers a fluency-logic dissociation: high invariance to surface-level style changes reward changes $<$0.1, yet inconsistent detection of logically-corrupted reasoning, with different models failing on different attack types. Adversarial optimization demonstrates that gradient-based attacks inflate rewards on invalid trajectories, with reward landscapes exhibiting wide, exploitable peaks. RL-induced reward hacking exposes the critical failure mode: policies trained on AIME problems achieve near-perfect PRM rewards ($>$0.9), while ground-truth accuracy remains low (below 4%), with 43% of reward gains attributable to stylistic shortcuts. These findings reveal that current PRMs function as fluency detectors rather than reasoning verifiers, creating systematic blind spots that undermine their use as training signals. We release PRM-BiasBench and a diagnostic toolkit to enable robustness evaluation before deployment. The code and dataset are available at https://github.com/SqueezeAILab/reward-under-attack.

📄 PDF Abstract BibTeX arXiv:2603.06621

Code (0)

등록된 구현이 없습니다.

Similar Papers 제목 키워드 기반

Defining and Characterizing Reward Hacking

2022-09-27 · Joar Skalse, Nikolaus H. R. Howe, Dmitrii Krasheninnikov, David Krueger

We provide the first formal definition of reward hacking, a phenomenon where optimizing an imperfect proxy reward function, $\mathcal{\tilde{R}}$, leads to poor performance according to the true reward function, $\mathca…

Robust Q-Learning under Corrupted Rewards

2024-09-05 · Sreejeet Maity, Aritra Mitra

Recently, there has been a surge of interest in analyzing the non-asymptotic behavior of model-free reinforcement learning algorithms. However, the performance of such algorithms in non-ideal environments, such as in the…

Q-Learning

Perturbing the Phase: Analyzing Adversarial Robustness of Complex-Valued Neural Networks

2026-02-06 · Florian Eilers, Christof Duhme, Xiaoyi Jiang arxiv

Complex-valued neural networks (CVNNs) are rising in popularity for all kinds of applications. To safely use CVNNs in practice, analyzing their robustness against outliers is crucial. One well known technique to understa…

Adversarial Robustness

Imperfect World Models are Exploitable

2026-05-15 · Logan Mondal Bhamidipaty, Esmeralda S. Whitammer, David Abel, Mykel J. Kochenderfer 외 arxiv

We propose a novel definition of model exploitation in reinforcement learning. Informally, a world model is exploitable if it implies that one policy should be strictly preferred over another while the environment's true…

Reinforcement Learning

On the Robustness of Safe Reinforcement Learning under Observational Perturbations

2022-05-29 · Zuxin Liu, Zijian Guo, Zhepeng Cen, huan zhang 외

Safe reinforcement learning (RL) trains a policy to maximize the task reward while satisfying safety constraints. While prior works focus on the performance optimality, we find that the optimal solutions of many safe RL …

Adversarial Attackreinforcement-learningReinforcement LearningReinforcement Learning (RL)+1