相关论文: Evaluating Evidence Grounding Under User Pressure …

Large language models (LLMs) are known to abandon their initial stance to conform to user pushback. While prior research largely attributes this behavior to sycophancy learned during reinforcement learning from human feedback, we…

Large language models are widely adopted as automated evaluation judges, yet the stability of their verdicts under semantically equivalent prompt rephrasings remains largely unexamined. We conduct a systematic empirical study of…

Modern language models (LMs) need to follow human instructions while being faithful; yet, they often fail to achieve both. Here, we provide concrete evidence of a trade-off between instruction following (i.e., follow open-ended…

Large language models are increasingly relied upon as sources of information, but their propensity for generating false or misleading statements with high confidence poses risks for users and society. In this paper, we confront the critical…

Instruction following is a critical ability for Large Language Models to perform downstream tasks. The standard approach to instruction tuning has relied on a specific phase of supervised fine-tuning over curated instruction datasets,…

Large language models (LLMs) are increasingly used as automatic judges for summarization and dialogue evaluation. Prior work has documented biases such as position, verbosity, and style preferences, but largely focuses on outcomes, leaving…

Longitudinal NLP tasks require reasoning over temporally ordered text to detect persistence and change in human behavior and opinions. However, in-context learning with large language models struggles on tasks where models must integrate…

Language Models (LMs) continue to advance, improving response quality and coherence. Given Internet-scale training datasets, LMs have likely encountered much of what users may ask them to generate in some form during their training. A…

This paper presents a comprehensive exploration of leveraging Large Language Models (LLMs), specifically GPT-4, in the field of instructional design. With a focus on scaling evidence-based instructional design expertise, our research aims…

Instruction tuning is a widely used approach to improve the instruction-following ability of large language models (LLMs). Instruction-tuning datasets typically include a mixture of context-augmented and context-free examples, yet prior…

Most large language models (LLMs) are sensitive to prompts, and another synonymous expression or a typo may lead to unexpected results for the model. Composing an optimal prompt for a specific demand lacks theoretical support and relies…

Sycophancy, the tendency of language models to prioritize agreement with user preferences over principled reasoning, has been identified as a persistent alignment failure in English-language evaluations. However, it remains unclear whether…

Code generation models are widely used in software development, yet their sensitivity to prompt phrasing remains under-examined. Identical requirements expressed with different emotions or communication styles can yield divergent outputs,…

Grounding has been argued to be a crucial component towards the development of more complete and truly semantically competent artificial intelligence systems. Literature has divided into two camps: While some argue that grounding allows for…

Large Language Models (LLMs) are highly vulnerable to input perturbations, as even a small prompt change may result in a substantially different output. Existing methods to enhance LLM robustness are primarily focused on perturbed data…

Large language models leverage both parametric knowledge acquired during pretraining and in-context knowledge provided at inference time. Crucially, when these sources conflict, models arbitrate based on their internal confidence,…

Language models deployed in high-stakes professional settings face conflicting demands from users, institutional authorities, and professional norms. How models act when these demands conflict reveals a principal hierarchy -- an implicit…

Automated fact-checking has been a challenging task for the research community. Prior work has explored various strategies, such as end-to-end training, retrieval-augmented generation, and prompt engineering, to build robust fact-checking…

Long-horizon interactions require language models to manage accumulating information: when to update their state, when to preserve their state, and what to ignore. We study this challenge as \textbf{Contextual Belief Management (CBM)}:…

Large language models (LLMs) play a key role in generating evidence-based and stylistic counter-arguments, yet their effectiveness in real-world applications has been underexplored. Previous research often neglects the balance between…