Detecting and Suppressing Reward Hacking with Gradient Fingerprints

• URL: http://arxiv.org/abs/2604.16242v1
• Date: Fri, 17 Apr 2026 17:01:24 GMT
• Title: Detecting and Suppressing Reward Hacking with Gradient Fingerprints
• Abstract summary: Fingerprint Gradient (GRIFT) is a method for detecting reward hacking using models' internal computations.<n>Across verifiable reasoning benchmarks, GRIFT substantially outperforms strong baselines.<n>Our results highlight a promising direction of leveraging gradient level representations for assessing the quality of CoT reasoning traces.
• Score: 46.25008147178368
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Reinforcement learning with verifiable rewards (RLVR) typically optimizes for outcome rewards without imposing constraints on intermediate reasoning. This leaves training susceptible to reward hacking, where models exploit loopholes (e.g., spurious patterns in training data) in the reward function to achieve high scores without solving the intended task. These reward-hacking behaviors are often implicit, as the intermediate chain-of-thought (CoT) may appear plausible on the surface, limiting the effectiveness of purely text-based monitoring. We propose Gradient Fingerprint (GRIFT), a method for detecting reward hacking using models' internal computations. Given a prompt and a model-generated CoT, GRIFT computes gradients of the CoT conditioned on the prompt and compresses them into a compact representation, which is then used to assess whether the CoT reflects reward hacking behavior. Across verifiable reasoning benchmarks spanning math, code, and logical reasoning, GRIFT substantially outperforms strong baselines, including CoT Monitor and TRACE, achieving over 25% relative improvement in detecting reward hacking behavior. Moreover, integrating GRIFT into the rejection fine-tuning pipeline for reasoning tasks reduces reward hacking and improves performance on the true task objective. Our results highlight a promising direction of leveraging gradient level representations for assessing the quality of CoT reasoning traces. Our code is available at: https://github.com/songtao-x/reward_hack.

Related papers

• Gradients Know What Outcomes Don't: Unlocking Reinforcement Learning for LLM Reasoning with Gradient-Aligned Rewards [24.636191382674184]
  Gradient Reward (GAR) operates in the policy's own gradient space.<n>C cosine similarity with an expert-anchor gradient yields a dense, reasoning-aware reward.
  arXiv  Detail & Related papers  (2026-09-03T04:00:28Z)
• VeriGate: Verifier-Gated Step-Level Supervision for GRPO [51.26100506256885]
  Group Relative Policy Optimization is an effective recipe for training reasoning models with verifier-based outcome rewards.<n>We present VeriGate, a verifier-gated extension of GRPO that addresses these limitations with three design choices.<n>We show that VeriGate improves average accuracy by about 20% and 12% for 1.5B and 7B models respectively.
  arXiv  Detail & Related papers  (2026-05-28T18:20:32Z)
• Verifier-Free RL for LLMs via Intrinsic Gradient-Norm Reward [69.99652051809737]
  In this work, we propose Verifier-free Intrinsic Gradient-Norm Reward (VIGOR)<n>VIGOR is a simple reward that uses only the policy model itself.<n>It exhibits cross-domain transfer to code benchmarks when trained only on math data.
  arXiv  Detail & Related papers  (2026-05-11T03:15:37Z)
• Do Synthetic Trajectories Reflect Real Reward Hacking? A Systematic Study on Monitoring In-the-Wild Hacking in Code Generation [53.024513172383195]
  We present a systematic analysis of the synthetic vs. in-the-wild discrepancy in reward hacking.<n>We examine what extent hacking behaviors induced by prompting resemble those emerging during RL training.<n>We find that synthetic-data-trained monitors fail to generalize to "in-the-wild" hacking.
  arXiv  Detail & Related papers  (2026-04-26T01:26:50Z)
• When Reward Hacking Rebounds: Understanding and Mitigating It with Representation-Level Signals [11.280037154530847]
  Reinforcement learning for LLMs is vulnerable to reward hacking.<n>We study this phenomenon in coding tasks using an environment-manipulation setting.<n>We propose Advantage Modification, which integrates shortcut concept scores into GRPO advantage computation.
  arXiv  Detail & Related papers  (2026-04-01T23:33:08Z)
• Countdown-Code: A Testbed for Studying The Emergence and Generalization of Reward Hacking in RLVR [15.115899490498341]
  Reward hacking is a form of misalignment in which models overoptimize proxy rewards without genuinely solving the underlying task.<n>We introduce Countdown-Code, a minimal environment where models can both solve a mathematical reasoning task and manipulate the test harness.<n>We study reward hacking in open-weight LLMs and find that such behaviors can be unintentionally learned during supervised fine-tuning.
  arXiv  Detail & Related papers  (2026-03-07T07:43:14Z)
• IR$^3$: Contrastive Inverse Reinforcement Learning for Interpretable Detection and Mitigation of Reward Hacking [67.20568716300272]
  Reinforcement Learning from Human Feedback (RLHF) enables powerful LLM alignment but can introduce reward hacking.<n>We introduce IR3 (Interpretable Reward Reconstruction and Rectification), a framework that reverse-engineers, interprets, and surgically repairs the implicit objectives driving RLHF-tuned models.<n>We show that IR3 achieves 0.89 correlation with ground-truth rewards, identifies hacking features with over 90% precision, and significantly reduces hacking behaviors while maintaining capabilities within 3% of the original model.
  arXiv  Detail & Related papers  (2026-02-23T01:14:53Z)
• Adversarial Reward Auditing for Active Detection and Mitigation of Reward Hacking [69.06218054848803]
  We propose Adrial Reward Auditing (ARA), a framework that reconceptualizes reward hacking as a dynamic, competitive game.<n>ARA operates in two stages: first, a Hacker policy discovers reward model vulnerabilities while an Auditor learns to detect exploitation from latent representations.<n>ARA achieves the best alignment-utility tradeoff among all baselines.
  arXiv  Detail & Related papers  (2026-02-02T07:34:57Z)
• From Verifiable Dot to Reward Chain: Harnessing Verifiable Reference-based Rewards for Reinforcement Learning of Open-ended Generation [52.62655622099456]
  We propose reinforcement learning with verifiable reference-based rewards (RLVRR)<n>Instead of checking the final answer, RLVRR extracts an ordered linguistic signal from high-quality references (i.e., reward chain)<n>In this way, RLVRR decomposes rewards into two dimensions: content, which preserves deterministic core concepts, and style, which evaluates adherence to stylistic properties.
  arXiv  Detail & Related papers  (2026-01-26T14:39:58Z)
• Is It Thinking or Cheating? Detecting Implicit Reward Hacking by Measuring Reasoning Effort [44.34183850072512]
  Reward hacking, where a reasoning model exploits loopholes in a reward function to achieve high rewards without solving the intended task, poses a significant threat.<n>To detect implicit reward hacking, we propose TRACE (Truncated Reasoning AUC Evaluation)<n>Our key observation is that hacking occurs when exploiting the loophole is easier than solving the actual task.
  arXiv  Detail & Related papers  (2025-10-01T18:49:45Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.