相关论文: Steerable Pluralism: Pluralistic Alignment via Few…

The ability of LLMs to represent diverse perspectives is critical as they increasingly impact society. However, recent studies reveal that alignment algorithms such as RLHF and DPO significantly reduce the diversity of LLM outputs. Not only…

The prevailing approach to aligning Large Language Models (LLMs) typically relies on human or AI feedback and assumes access to specific types of preference datasets. In our work, we question the efficacy of such datasets and explore…

Performance of Large Language Models (LLMs) on multiple-choice tasks differs markedly between symbol-based and cloze-style evaluation formats. The observed discrepancies are systematically attributable to task characteristics: natural…

Recently, tremendous strides have been made to align the generation of Large Language Models (LLMs) with human values to mitigate toxic or unhelpful content. Leveraging Reinforcement Learning from Human Feedback (RLHF) proves effective and…

Deep Reinforcement Learning is widely used for aligning Large Language Models (LLM) with human preference. However, the conventional reward modelling is predominantly dependent on human annotations provided by a select cohort of…

Large Language Models (LLMs) exhibit impressive capabilities but require careful alignment with human preferences. Traditional training-time methods finetune LLMs using human preference datasets but incur significant training costs and…

Reward Models (RMs) are crucial for online alignment of language models (LMs) with human preferences. However, RM-based preference-tuning is vulnerable to reward hacking, whereby LM policies learn undesirable behaviors from flawed RMs. By…

Few-shot learning aims to train models that can be generalized to novel classes with only a few samples. Recently, a line of works are proposed to enhance few-shot learning with accessible semantic information from class names. However,…

The success of AI assistants based on Language Models (LLMs) hinges on Reinforcement Learning from Human Feedback (RLHF) to comprehend and align with user intentions. However, traditional alignment algorithms, such as PPO, are hampered by…

Large Language Models (LLMs), when used in educational settings without pedagogical fine-tuning, often provide immediate answers rather than guiding students through the problem-solving process. This approach falls short of pedagogically…

Reinforcement Learning frameworks, particularly those utilizing human annotations, have become an increasingly popular method for preference fine-tuning, where the outputs of a language model are tuned to match a certain set of behavioral…

Preference learning is critical for aligning large language models (LLMs) with human values, with the quality of preference datasets playing a crucial role in this process. While existing metrics primarily assess data quality based on…

Large language models (LLMs) are increasingly used as automated judges to evaluate recommendation systems, search engines, and other subjective tasks, where relying on human evaluators can be costly, time-consuming, and unscalable. LLMs…

Multi-objective alignment aims to align LLM responses with multiple human preference objectives. Among existing methods, guiding the generation of frozen LLMs through autoregressive reward models (ARMs) to accomplish multi-objective…

Aligning human preference and value is an important requirement for building contemporary foundation models and embodied AI. However, popular approaches such as reinforcement learning with human feedback (RLHF) break down the task into…

Mainstream approaches to aligning large language models (LLMs) heavily rely on human preference data, particularly when models require periodic updates. The standard process for iterative alignment of LLMs involves collecting new human…

We study the problem of computing an optimal large language model (LLM) policy for the constrained alignment problem, where the goal is to maximize a primary reward objective while satisfying constraints on secondary utilities. Despite the…

Reinforcement Learning from Human Feedback (RLHF) is a key method for aligning large language models (LLMs) with human preferences. However, current offline alignment approaches like DPO, IPO, and SLiC rely heavily on fixed preference…

Alignment techniques have become central to ensuring that Large Language Models (LLMs) generate outputs consistent with human values. However, existing alignment paradigms often model an averaged or monolithic preference, failing to account…

Reinforcement learning with verifiable rewards (RLVR) has achieved remarkable success in logical reasoning tasks, yet whether large language model (LLM) alignment requires fundamentally different approaches remains unclear. Given the…