arXiv Machine Learning
Towards Bridging the Gap Between Offline and Iterative Alignment via Preference Distillation
Read the original on arXiv Machine Learning →The Flow has not summarised this story yet — read it at arXiv Machine Learning.
Related stories
arXiv AI
1d ago
A Zeroth-Order Paradigm for LLM Preference Alignment
The paper introduces Comparison-based Preference Optimization (ComPO), a zeroth-order method that aligns large language models with human preferences using comparison oracles instead of direct differentiable loss optimization. It provides theoretical convergence guarantees for both offline and online variants under smoothness, gradient sparsity, and oracle compatibility assumptions, and establishes performance bounds under local coverage and in-distribution reward accuracy. Experiments on several LLMs (Mistral, Llama, Gemma-2, Qwen3, Gemma-3) show that ComPO outperforms existing direct alignment methods, achieving higher length-controlled win rates and diagnostics that suggest mitigation of likelihood displacement.
arXiv AI
Aug 18
Leveraging Machine Unlearning for Cost-Efficient Preference Alignment
arXiv:2504. 06659v2 Announce Type: replace-cross Abstract: Despite advances in Preference Alignment (PA) for Large Language Models (LLMs), mainstream methods like reinforcement learning with human feedback face notable challenges.
arXiv AI
Jun 11
Toward Preference-aligned Large Language Models via Residual-based Model Steering
arXiv:2509. 23982v2 Announce Type: replace-cross Abstract: Preference alignment is a critical step in making Large Language Models (LLMs) useful and aligned with (human) preferences.
arXiv AI
Jun 10
A Comprehensive Survey of Direct Preference Optimization: Datasets, Theories, Variants, and Applications
arXiv:2410. 15595v4 Announce Type: replace Abstract: With the rapid advancement of large language models (LLMs), aligning policy models with human preferences has become increasingly critical.
arXiv AI
Aug 26
Preference Data Selection for Mitigating the Alignment Tax in Large Language Models
The paper introduces BALIGN, a balanced data selection strategy designed to reduce catastrophic forgetting—referred to as the alignment tax—in large language models during preference-based alignment. By analyzing preference optimization gradients, the authors identify three data-centric features that influence parameter drift: the reference model's log-probability margin, token length differences between chosen and rejected responses, and TF‑IDF similarity to general capability corpora. BALIGN aggregates these features into a composite risk score to filter out high-risk preference samples, thereby preserving foundational capabilities while maintaining alignment gains with minimal computational overhead.
Hugging Face Trending Papers
Aug 25
Preference Data Selection for Mitigating the Alignment Tax in Large Language Models
Aligning large language models to human preferences is crucial for real-world deployment but frequently incurs an alignment tax, leading to the catastrophic forgetting of pre-trained general capabilit...