LLM Alignment

LLM: Large Language Model

Overview

LLM alignment studies how to shape language models’ behavior so they are helpful, truthful, safe, and responsive to the people and settings they serve. Common approaches include supervised fine-tuning, preference learning such as Direct Preference Optimization, reinforcement learning from human or AI feedback, and inference-time steering or retrieval. Researchers evaluate these methods on instruction-following, preference, jailbreak, factuality, multilingual, and multi-turn tasks, while also developing ways to represent disagreement among users and communities. A central challenge is that gains in safety or preference alignment may reduce usefulness, diversity, calibration, or capability, and desired behavior may not generalize to new prompts, groups, or deployment settings.

Latest papers

634 papers24 papers in the last four weeks, down 64% on the four weeks before. 0.2% of all new papers.

Jan 16, 2025

Foundations of Large Language Models

This is a book about large language models. As indicated by the title, it primarily focuses on foundational concepts rather than comprehensive coverage of all cutting-edge technologies. The book is structured into six main chapters, each exploring a key area: pre-training, generative models, prompting, alignment, inference, and reasoning. It is intended for college students, professionals, and practitioners in natural language processing and related fields, and can serve as a reference for anyone interested in large language models.
Oct 31, 2024

Risk-Aware Alignment Must Incorporate Reward Uncertainty

We consider the problem of policy optimization with noisy rewards, where the policy is constrained to be in the neighborhood of a reference policy.We study the risk of the policy optimization process, defined to be the chance of obtaining a policy worse off than the reference policy. In an idealized multivariate Gaussian model for the noisy reward function, we show that the regularizer achieving optimal risk in a suitably local neighborhood of the reference policy is the quadratic norm weighted by the reward covariance matrix. This result is used to develop a new policy optimization algorithm based on regularization via reward variance estimates. Experiments across synthetic bandit models, large language models and reward models show that this algorithm yields more robust alignment outcomes than existing approaches.
Jun 3, 2024

Decoupled Alignment for Robust Plug-and-Play Adaptation

We introduce a training-free safety enhancement method for aligning large language models (LLMs) without the need for supervised fine-tuning or reinforcement learning from human feedback. Our main idea is to provide a robust plug-and-play approach to prevent shadow alignment when models are adapted to downstream tasks. Specifically, we leverage knowledge distillation to extract alignment signals from well-aligned LLMs and inject them into shadow-aligned models via model fusion, enabling plug-and-play alignment correction. In our methodology, we employ delta debugging to identify the critical components of knowledge necessary for effective distillation. On the harmful question dataset, our method significantly enhances the average defense success rate by approximately 14.42%, reaching as high as 51.39% across 17 influenced LLMs, without compromising performance. Our code is available at https://github.com/NWULIST/DAPA.
Feb 2, 2024

KTO: Model Alignment as Prospect Theoretic Optimization

Kahneman & Tversky's prospect theory\textit{prospect theory} tells us that humans perceive random variables in a biased but well-defined manner (1992); for example, humans are famously loss-averse. We show that objectives for aligning LLMs with human feedback implicitly incorporate many of these biases -- the success of these objectives (e.g., DPO) over cross-entropy minimization can partly be ascribed to them belonging to a family of loss functions that we call human-aware losses\textit{human-aware losses} (HALOs). However, the utility functions these methods attribute to humans still differ from those in the prospect theory literature. Using a Kahneman-Tversky model of human utility, we propose a HALO that directly maximizes the utility of generations instead of maximizing the log-likelihood of preferences, as current methods do. We call this approach KTO, and it matches or exceeds the performance of preference-based methods at scales from 1B to 30B, despite only learning from a binary signal of whether an output is desirable. More broadly, our work suggests that there is no one HALO that is universally superior; the best loss depends on the inductive biases most appropriate for a given setting, an oft-overlooked consideration.