Preference Instability in Reward Models: Detection and Mitigation via Sparse Autoencoders
Authors: Shunchang Liu, Xin Chen, Belen Martin Urcelay, Francesco Croce
Organizations: ETH Zürich · Georgia Institute of Technology · ELLIS Institute Finland · Aalto University
Abstract
Preference learning in large language models relies on reward models as proxies for human judgment. However, these models frequently exhibit preference instability, producing contradictory preference assignments in response to subtle, meaning-preserving input variations. We analyze this instability at the representation level under three semantic-preserving perturbation types: paraphrasing, pattern injection, and backdoor triggers. We attribute this instability to over-reliance on predictive yet brittle features, which we term unstable features, and isolate them via Sparse Autoencoders (SAEs) in a sparse latent space where benign and perturbed inputs activate distinctly separable patterns. Building on this separability, we propose two SAE-based instability mitigation strategies: SAE Feature Steering, which identifies and suppresses anomalously activated features at inference, and SAE Residual Correction, which learns adaptive adjustments over SAE features to restore correct preferences. Our methods substantially reduce incorrect preference assignments on harmlessness and hallucination benchmarks while preserving benign performance and general utility on other tasks, without retraining the reward model. Our code and data are available in \url{https://github.com/shunchang-liu/pisa}.
Reward models (RMs) play a central role in aligning large language models (LLMs) with human preferences. However, RMs are often sensitive to spurious features such as response length. Existing inference-time approaches for mitigating these biases typically focus exclusively on response length, resulting in performance trade-offs. In this paper, we propose causally motivated intervention for mitigating multiple types of biases in RMs at inference time. Our method first identifies neurons whose activations are strongly correlated with predefined bias attributes, and applies neuron-level intervention that suppresses these signals. We evaluate our method on RM benchmarks and observe reductions in sensitivity to spurious features across diverse bias types, without inducing performance trade-offs. Moreover, when used for preference annotation, small RMs (2B and 7B) with our method, which edits less than 2% of all the neurons in RMs, enable LLMs to improve alignment, achieving performance comparable to that of a state-of-the-art 70B RM on AlpacaEval and MT-Bench. Further analysis reveals that bias signals are primarily encoded by neurons in early layers, shedding light on the internal mechanisms of bias exploitation in RMs.
The alignment of large language models with human preferences is commonly achieved through Reinforcement Learning from Human Feedback or Direct Preference Optimization. However, these methods are vulnerable to the significant noise prevalent in real-world preference datasets. To address this critical issue, we present a theoretical framework for unbiased alignment, introducing the Unbiased Reward Model (URM) loss and the Unbiased Direct Preference Optimization (UDPO) loss. By mathematically correcting the distortion induced by preference noise, our novel objectives enable unbiased model training directly from noisy datasets, without requiring clean ground-truth supervision. We provide rigorous theoretical analyses demonstrating that our methods are noise-tolerant, parameter downward compatible, and classification-calibrated. Comprehensive experiments across diverse datasets demonstrate that our approaches outperform state-of-the-art baselines. Code available at: https://github.com/cswjl/unbiased-alignment.
Large language models (LLMs) increasingly serve as automated evaluators, yet they suffer from "self-preference bias": a tendency to favor their own outputs over those of other models. This bias undermines fairness and reliability in evaluation pipelines, particularly for tasks like preference tuning and model routing. We investigate whether lightweight steering vectors can mitigate this problem at inference time without retraining. We introduce a curated dataset that distinguishes self-preference bias into justified examples of self-preference and unjustified examples of self-preference, and we construct steering vectors using two methods: Contrastive Activation Addition (CAA) and an optimization-based approach. Our results show that steering vectors can reduce unjustified self-preference bias by up to 97%, substantially outperforming prompting and direct preference optimization baselines. Yet steering vectors are unstable on legitimate self-preference and unbiased agreement, implying self-preference spans multiple or nonlinear directions. This underscores both their promise and limits as safeguards for LLM-as-judges and motivates more robust interventions.
Dani Roytburg, Matthew Bozoukov, Matthew Nguyen +3