wDPO: Winsorized Direct Preference Optimization for Robust LLM Alignment

• URL: http://arxiv.org/abs/2603.07211v1
• Date: Sat, 07 Mar 2026 13:30:53 GMT
• Title: wDPO: Winsorized Direct Preference Optimization for Robust LLM Alignment
• Abstract summary: In practice, preference data are often noisy.<n>Existing robust variants of DPO mainly rely on uniform objective modifications or global reweighting.<n>We show that robust preference alignment benefits from addressing different noise types with targeted interventions.
• Score: 48.487557157323664
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Direct Preference Optimization (DPO) aligns large language models by optimizing pairwise preferences and has shown remarkable effectiveness as a simple and scalable alternative to RLHF. However, in practice, preference data are often noisy. Existing robust variants of DPO mainly rely on uniform objective modifications or global reweighting. While partially effective, these methods treat noisy samples as a homogeneous source of uncertainty and fail to distinguish between different noise types, leading to sub-optimal alignment robustness. In this work, we show that robust preference alignment benefits from addressing different noise types with targeted interventions rather than uniform regularization. We propose winsorized Direct Preference Optimization~(wDPO), a robust LLM alignment approach with hierarchical winsorization. Specifically, wDPO adopts a reward-free hierarchical intervention strategy that leverages only signals already available during DPO training. It first uses the implicit margin from DPO log-ratio to identify heterogeneous noise patterns without relying on external reward models. For hard noise, wDPO performs a data-level intervention by sparsely correcting strongly inconsistent preference pairs. For ambiguous comparisons, it applies a gradient-level intervention through soft winsorization, capping extreme losses in the high-loss tail to prevent weakly informative samples from dominating gradient updates. Extensive experiments on PKU-SafeRLHF and multiple external safety benchmarks demonstrate that wDPO consistently improves preference alignment quality and robustness over vanilla DPO and strong DPO-family baselines, with particularly pronounced gains under controlled label-flip noise.

Related papers

• Private Direct Preference Optimization for LLM Alignment [77.56374641532715]
  Direct preference optimization (DPO) is now a standard method for aligning large language models (LLMs) using human preference data.<n>In this paper, we formalize preference privacy, a label-DP-style privacy notion for DPO.<n>We then design PrivDPO, a DPO variant that enforces preference privacy while remaining compatible with large-scale LLM training.
  arXiv  Detail & Related papers  (2026-08-05T16:51:02Z)
• AdaDPO: Self-Adaptive Direct Preference Optimization with Balanced Gradient Updates [0.03999851878220877]
  We propose a Self-Adaptive variant of the DPO algorithm that introduces per-preference-pair, stop-gradient-based coefficients.<n>AdaDPO is constructed to enforce equality of gradient magnitudes between preferred and dispreferred probabilities.<n>Because it operates purely at the loss level, AdaDPO can be dropped into existing preference-based alignment pipelines.
  arXiv  Detail & Related papers  (2026-05-27T13:05:49Z)
• Conditional Equivalence of DPO and RLHF: Implicit Assumption, Failure Modes, and Provable Alignment [51.18269946911088]
  Direct Preference Optimization (DPO) has emerged as a popular alternative to Reinforcement Learning from Human Feedback (RLHF)<n>We show that DPO and RLHF optimize fundamentally different objectives in such cases.<n>We introduce Constrained Preference Optimization (CPO), augmenting RLHF with constraints for provable alignment.<n>Our theoretical analysis establishes when DPOs' guarantees hold and provides solutions preserving simplicity with provable alignment.
  arXiv  Detail & Related papers  (2026-05-20T07:26:22Z)
• Lightweight Robust Direct Preference Optimization [26.99327564250612]
  We propose DPO-PRO (DPO with Preference Robustness), a robust fine-tuning algorithm based on DPO.<n>Unlike prior DRO-based variants, DPO-PRO focuses solely on uncertainty in preferences, avoiding unnecessary conservatism and incurring negligible computational overhead.
  arXiv  Detail & Related papers  (2025-10-27T17:55:06Z)
• Margin Adaptive DPO: Leveraging Reward Model for Granular Control in Preference Optimization [0.0]
  Margin-Adaptive Direct Preference Optimization provides a stable, data-preserving, and instance-level solution.<n>We provide a comprehensive theoretical analysis, proving that MADPO has a well-behaved optimization landscape.<n>It achieves performance gains of up to +33.3% on High Quality data and +10.5% on Low Quality data over the next-best method.
  arXiv  Detail & Related papers  (2025-10-06T20:09:37Z)
• Inducing Robustness in a 2 Dimensional Direct Preference Optimization Paradigm [16.66633426354087]
  Direct Preference optimisation (DPO) has emerged as a powerful method for aligning Large Language Models with human preferences.<n>We investigate the performance of DPO using open-source preference datasets.<n>We propose an approach of incorporating segment-level score noise robustness to the 2D-DPO algorithm.
  arXiv  Detail & Related papers  (2025-05-03T05:59:13Z)
• Uncertainty-Penalized Direct Preference Optimization [52.387088396044206]
  We develop a pessimistic framework for DPO by introducing preference uncertainty penalization schemes. The penalization serves as a correction to the loss which attenuates the loss gradient for uncertain samples. We show improved overall performance compared to vanilla DPO, as well as better completions on prompts from high-uncertainty chosen/rejected responses.
  arXiv  Detail & Related papers  (2024-10-26T14:24:37Z)
• Towards Robust Alignment of Language Models: Distributionally Robustifying Direct Preference Optimization [45.6430987775264]
  This study addresses the challenge of noise in training datasets for Direct Preference Optimization (DPO)<n>We categorize noise into pointwise noise, which includes low-quality data points, and pairwise noise, which encompasses erroneous data pair associations.<n>We introduce Distributionally Robustifying DPO, which integrates pairwise robustness by optimizing against worst-case pairwise scenarios.
  arXiv  Detail & Related papers  (2024-07-10T17:48:25Z)
• Provably Mitigating Overoptimization in RLHF: Your SFT Loss is Implicitly an Adversarial Regularizer [52.09480867526656]
  We identify the source of misalignment as a form of distributional shift and uncertainty in learning human preferences.<n>To mitigate overoptimization, we first propose a theoretical algorithm that chooses the best policy for an adversarially chosen reward model.<n>Using the equivalence between reward models and the corresponding optimal policy, the algorithm features a simple objective that combines a preference optimization loss and a supervised learning loss.
  arXiv  Detail & Related papers  (2024-05-26T05:38:50Z)
• ROPO: Robust Preference Optimization for Large Language Models [59.10763211091664]
  We propose an iterative alignment approach that integrates noise-tolerance and filtering of noisy samples without the aid of external models. Experiments on three widely-used datasets with Mistral-7B and Llama-2-7B demonstrate that ROPO significantly outperforms existing preference alignment methods.
  arXiv  Detail & Related papers  (2024-04-05T13:58:51Z)

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.