中文
相关论文

相关论文: Think Before You Lie: How Reasoning Leads to Hones…

200 篇论文

Large language models (LLMs) often improve their performance in downstream tasks when they generate Chain of Thought reasoning text before producing an answer. We investigate how LLMs recover from errors in Chain of Thought. Through…

人工智能 · 计算机科学 2024-09-04 Evelyn Yee , Alice Li , Chenyu Tang , Yeon Ho Jung , Ramamohan Paturi , Leon Bergen

Unlike autoregressive models, which generate tokens sequentially and benefit from reasoning-before-answering strategies such as Chain-of-Thought, Masked Diffusion Language Models (MDLMs) refine all sequence positions simultaneously, raising…

计算与语言 · 计算机科学 2026-03-03 Jacob Devasier

The recent trend towards utilisation of reasoning models has improved the performance of Large Language Models (LLMs) across many tasks which involve logical steps. One linguistic task that could benefit from this framing is idiomaticity…

计算与语言 · 计算机科学 2025-08-20 Dylan Phelps , Rodrigo Wilkens , Edward Gow-Smith , Thomas Pickard , Maggie Mi , Aline Villavicencio

Large Language Models (LLMs) have been shown to achieve impressive results for many reasoning-based NLP tasks, suggesting a degree of deductive reasoning capability. However, it remains unclear to which extent LLMs, in both informal and…

计算与语言 · 计算机科学 2025-08-26 Fabian Hoppe , Filip Ilievski , Jan-Christoph Kalo

The development of Large Language Models (LLMs) has notably transformed numerous sectors, offering impressive text generation capabilities. Yet, the reliability and truthfulness of these models remain pressing concerns. To this end, we…

计算与语言 · 计算机科学 2024-02-12 Satyapriya Krishna , Chirag Agarwal , Himabindu Lakkaraju

Large Language Models (LLMs) have been demonstrating strong reasoning capability with their chain-of-thoughts (CoT), which are routinely used by humans to judge answer quality. This reliance creates a powerful yet fragile basis for trust.…

机器学习 · 计算机科学 2026-05-22 Wei Shen , Han Wang , Haoyu Li , Huan Zhang

Large Language Models (LLMs) are increasingly positioned as decision engines for hiring, healthcare, and economic judgment, yet real-world human judgment reflects a balance between rational deliberation and emotion-driven bias. If LLMs are…

In this paper, we investigate the degree to which fine-tuning in Large Language Models (LLMs) effectively mitigates versus merely conceals undesirable behavior. Through the lens of semi-realistic role-playing exercises designed to elicit…

计算与语言 · 计算机科学 2024-07-01 Florin Pop , Judd Rosenblatt , Diogo Schwerz de Lucena , Michael Vaiana

We explore the ability of large language models (LLMs) to engage in subtle deception through strategically phrasing and intentionally manipulating information. This harmful behavior can be hard to detect, unlike blatant lying or…

Vision-language models (VLMs) are increasingly deployed in high-stakes settings where reliable uncertainty quantification (UQ) is as important as predictive accuracy. Extended reasoning via chain-of-thought (CoT) prompting or…

机器学习 · 计算机科学 2026-03-18 Robert Welch , Emir Konuk , Kevin Smith

Large language models (LLMs) have the potential to aid and improve human decision-making in classification tasks, not only by providing fairly accurate predictions, but also in their ability to generate cogent narrative explanations of…

人机交互 · 计算机科学 2026-05-25 Laura R. Marusich , Mary Grace Kozuch Dhooghe , Jonathan Z. Bakdash , Murat Kantarcioglu

Chain-of-thought explanations are widely used to inspect the decision process of large language models (LLMs) and to evaluate the trustworthiness of model outputs, making them important for effective collaboration between LLMs and humans.…

计算与语言 · 计算机科学 2025-07-16 Pedro Ferreira , Wilker Aziz , Ivan Titov

Large Language Models (LLMs) have impressive capabilities, but are prone to outputting falsehoods. Recent work has developed techniques for inferring whether a LLM is telling the truth by training probes on the LLM's internal activations.…

人工智能 · 计算机科学 2024-08-20 Samuel Marks , Max Tegmark

Do large language models (LLMs) display rational reasoning? LLMs have been shown to contain human biases due to the data they have been trained on; whether this is reflected in rational reasoning remains less clear. In this paper, we answer…

计算与语言 · 计算机科学 2024-02-16 Olivia Macmillan-Scott , Mirco Musolesi

Large Language Models (LLMs) have demonstrated strong generalization across a wide range of tasks. Reasoning with LLMs is central to solving multi-step problems and complex decision-making. To support efficient reasoning, recent studies…

计算与语言 · 计算机科学 2025-09-03 Jindong Li , Yali Fu , Li Fan , Jiahong Liu , Yao Shu , Chengwei Qin , Menglin Yang , Irwin King , Rex Ying

Large language models (LLMs) have demonstrated impressive capabilities in various reasoning tasks, aided by techniques like chain-of-thought prompting that elicits verbalized reasoning. However, LLMs often generate text with obvious…

人工智能 · 计算机科学 2024-12-06 Zhihui Xie , Jizhou Guo , Tong Yu , Shuai Li

Large language models (LLMs) are currently at the forefront of intertwining artificial intelligence (AI) systems with human communication and everyday life. Thus, aligning them with human values is of great importance. However, given the…

计算与语言 · 计算机科学 2024-06-06 Thilo Hagendorff

Large reasoning models (LRMs) have recently demonstrated impressive capabilities in complex reasoning tasks by leveraging increased test-time computation and exhibiting behaviors reminiscent of human-like self-reflection. While LRMs show a…

计算与语言 · 计算机科学 2025-10-21 Qingcheng Zeng , Weihao Xuan , Leyang Cui , Rob Voigt

While large language models (LLMs) excel in mathematical and code reasoning, we observe they struggle with social reasoning tasks, exhibiting cognitive confusion, logical inconsistencies, and conflation between objective world states and…

计算与语言 · 计算机科学 2025-10-14 Jialu Du , Guiyang Hou , Yihui Fu , Chen Wu , Wenqi Zhang , Yongliang Shen , Weiming Lu

Large language models (LLMs) can be dishonest when reporting on their actions and beliefs -- for example, they may overstate their confidence in factual claims or cover up evidence of covert actions. Such dishonesty may arise due to the…

机器学习 · 计算机科学 2025-12-24 Manas Joglekar , Jeremy Chen , Gabriel Wu , Jason Yosinski , Jasmine Wang , Boaz Barak , Amelia Glaese