D-STEER - Preference Alignment Techniques Learn to Behave, not to Believe -- Beneath the Surface, DPO as Steering Vector Perturbation in Activation Space
Direct Preference Optimization (DPO) has become a standard recipe for aligning large language models, yet it is still unclear what kind of change it actually induces inside the network. This paper argues that DPO does not rewrite a models internal beliefs; instead, it acts as a low rank steering mechanism that nudges activations along a small number of preference directions. Using a simple derivation, we show that the DPO gradient depends only on the difference between the logit embeddings of preferred and dispreferred completions, implying a first order shift in the final hidden representation rather than a deep restructuring of semantics. We then extract an empirical steering vector from a DPO tuned model and demonstrate that adding this vector to base activations reproduces most of the aligned behavior, while subtracting it nearly restores the original model. Finally, spectral analyses reveal rank-one dominance and entropy collapse in upper layers, indicating that alignment is funneled through a narrow subspace. Taken together, these results support a behavioral illusion view of DPO: it teaches models how to act aligned, not what to believe.
Code (0)
등록된 구현이 없습니다.
Similar Papers 제목 키워드 기반
Can Revealed Preferences Clarify LLM Alignment and Steering?
LLMs are increasingly used to make or support high-stakes decisions under uncertainty, where alignment depends not only on factual accuracy but on how models weigh tradeoffs between different outcomes. We present an empi…
Medical DiagnosisSteerable Pluralism: Pluralistic Alignment via Few-Shot Comparative Regression
Large language models (LLMs) are currently aligned using techniques such as reinforcement learning from human feedback (RLHF). However, these methods use scalar rewards that can only reflect user preferences on average. …
Reinforcement LearningDSPA: Dynamic SAE Steering for Data-Efficient Preference Alignment
Preference alignment is usually achieved by weight-updating training on preference data, which adds substantial alignment-stage compute and provides limited mechanistic visibility. We propose Dynamic SAE Steering for Pre…
The Anatomy of Alignment: Decomposing Preference Optimization by Steering Sparse Features
Prevailing alignment methods induce opaque parameter changes, obscuring what models truly learn. To address this, we introduce Feature Steering with Reinforcement Learning (FSRL), a framework that trains a lightweight ad…
Reinforcement LearningYaPO: Learnable Sparse Activation Steering Vectors for Domain Adaptation
Steering Large Language Models (LLMs) through activation interventions has emerged as a lightweight alternative to fine-tuning for alignment and personalization. Recent work on Bi-directional Preference Optimization (BiP…
General KnowledgeDomain Adaptation