相关论文: Revisiting Robustness for LLM Safety Alignment via…
Existing alignment methods for preference optimization of large language models (LLMs) aim to enhance model performance by utilizing pairs of positive and negative samples. However, due to the limited capacity of models in scoring or…
Supervised and preference-based fine-tuning techniques have become popular for aligning large language models (LLMs) with user intent and correctness criteria. However, real-world training data often exhibits spurious correlations --…
Post-training of LLMs with RLHF, and subsequently preference optimization algorithms such as DPO, IPO, etc., made a big difference in improving human alignment. However, all such techniques can only work with a single (human) objective. In…
Ensuring the safety of Large Language Models (LLMs) in diverse linguistic settings remains challenging, particularly for low-resource languages. Existing safety alignment methods are English-centric, limiting their effectiveness. We…
Direct Preference Optimization (DPO) is an efficient alignment technique that steers LLMs towards preferable outputs by training on preference data, bypassing the need for explicit reward models. Its simplicity enables easy adaptation to…
Though reasoning abilities are considered language-agnostic, existing LLMs exhibit inconsistent reasoning abilities across different languages, e.g., reasoning in the dominant language like English is superior to other languages due to the…
Aligning language models for both helpfulness and safety typically requires complex pipelines-separate reward and cost models, online reinforcement learning, and primal-dual updates. Recent direct preference optimization approaches simplify…
Large language models (LLMs) have shown promising abilities as cost-effective and reference-free evaluators for assessing language generation quality. In particular, pairwise LLM evaluators, which compare two generated texts and determine…
Large reasoning models (LRMs) generate intermediate reasoning traces before producing final answers, yielding strong gains on multi-step and mathematical tasks. Yet aligning LRMs with human preferences, a crucial prerequisite for model…
Aligning the output of Large Language Models (LLMs) with human preferences (e.g., by means of reinforcement learning with human feedback, or RLHF) is essential for ensuring their effectiveness in real-world scenarios. Despite significant…
In the field of large language models (LLMs), aligning models with the diverse preferences of users is a critical challenge. Direct Preference Optimization (DPO) has played a key role in this area. It works by using pairs of preferences…
An increasing number of NLP applications interact with large language models (LLMs) through black-box APIs, making prompt engineering critical for controlling model outputs. While recent Automatic Prompt Optimization (APO) methods…
We introduce ConfPO, a method for preference learning in Large Language Models (LLMs) that identifies and optimizes preference-critical tokens based solely on the training policy's confidence, without requiring any auxiliary models or…
Language model alignment methods such as reinforcement learning from human feedback (RLHF) have led to impressive advances in language model capabilities, but are limited by a widely observed phenomenon known as overoptimization, where the…
Lead optimization in drug discovery requires efficiently navigating vast chemical space through iterative cycles to enhance molecular properties while preserving structural similarity to the original lead compound. Despite recent advances,…
We propose Soft Preference Optimization (SPO), a method for aligning generative models, such as Large Language Models (LLMs), with human preferences, without the need for a reward model. SPO optimizes model outputs directly over a…
Fine-tuning large language models (LLMs) based on human preferences, commonly achieved through reinforcement learning from human feedback (RLHF), has been effective in improving their performance. However, maintaining LLM safety throughout…
Instruction following (IF) is a critical capability for large language models (LLMs). However, handling complex instructions with multiple constraints remains challenging. Previous methods typically select preference pairs based on the…
Off-policy updates are inevitable in reinforcement learning (RL) for large language models (LLMs) due to rollout staleness from asynchronous training and mismatches between training and inference engines. Naive importance sampling gives an…
While learning to align Large Language Models (LLMs) with human preferences has shown remarkable success, aligning these models to meet the diverse user preferences presents further challenges in preserving previous knowledge. This paper…