Aligning language models with human preferences
Language models (LMs) trained on vast quantities of text data can acquire sophisticated skills such as generating summaries, answering questions or generating code. However, they also manifest behaviors that violate human preferences, e.g., they can generate offensive content, falsehoods or perpetuate social biases. In this thesis, I explore several approaches to aligning LMs with human preferences. First, I argue that aligning LMs can be seen as Bayesian inference: conditioning a prior (base, pretrained LM) on evidence about human preferences (Chapter 2). Conditioning on human preferences can be implemented in numerous ways. In Chapter 3, I investigate the relation between two approaches to finetuning pretrained LMs using feedback given by a scoring function: reinforcement learning from human feedback (RLHF) and distribution matching. I show that RLHF can be seen as a special case of distribution matching but distributional matching is strictly more general. In chapter 4, I show how to extend the distribution matching to conditional language models. Finally, in chapter 5 I explore a different root: conditioning an LM on human preferences already during pretraining. I show that involving human feedback from the very start tends to be more effective than using it only during supervised finetuning. Overall, these results highlight the room for alignment techniques different from and complementary to RLHF.
Code (1)
Tasks
Bayesian InferenceSimilar Papers 제목 키워드 기반
Statistical Impossibility and Possibility of Aligning LLMs with Human Preferences: From Condorcet Paradox to Nash Equilibrium
Aligning large language models (LLMs) with diverse human preferences is critical for ensuring fairness and informed outcomes when deploying these models for decision-making. In this paper, we seek to uncover fundamental …
FairnessAligning Large Language Models with Human Preferences through Representation Engineering
Aligning large language models (LLMs) with human preferences is crucial for enhancing their utility in terms of helpfulness, truthfulness, safety, harmlessness, and interestingness. Existing methods for achieving this al…
Aligning Language Model Benchmarks with Pairwise Preferences
Language model benchmarks are pervasive and computationally-efficient proxies for real-world performance. However, many recent works find that benchmarks often fail to predict real utility. Towards bridging this gap, we …
Aligning Large Language Models with Implicit Preferences from User-Generated Content
Learning from preference feedback is essential for aligning large language models (LLMs) with human values and improving the quality of generated responses. However, existing preference learning methods rely heavily on c…
On the Algorithmic Bias of Aligning Large Language Models with RLHF: Preference Collapse and Matching Regularization
Accurately aligning large language models (LLMs) with human preferences is crucial for informing fair, economically sound, and statistically efficient decision-making processes. However, we argue that reinforcement learn…
Decision MakingText Generation