Large Language Models (LLMs) can achieve strong performance on many tasks by producing step-by-step reasoning before giving a final output, often referred to as chain-of-thought reasoning (CoT). It is tempting to interpret these CoT explanations as the LLM's process for solving a task. This level of transparency into LLMs' predictions would yield significant safety benefits. However, we find that CoT explanations can systematically misrepresent the true reason for a model's prediction. We demonstrate that CoT explanations can be heavily influenced by adding biasing features to model inputs--e.g., by reordering the multiple-choice options in a few-shot prompt to make the answer always "(A)"--which models systematically fail to mention in their explanations. When we bias models toward incorrect answers, they frequently generate CoT explanations rationalizing those answers. This causes accuracy to drop by as much as 36% on a suite of 13 tasks from BIG-Bench Hard, when testing with GPT-3.5 from OpenAI and Claude 1.0 from Anthropic. On a social-bias task, model explanations justify giving answers in line with stereotypes without mentioning the influence of these social biases. Our findings indicate that CoT explanations can be plausible yet misleading, which risks increasing our trust in LLMs without guaranteeing their safety. Building more transparent and explainable systems will require either improving CoT faithfulness through targeted efforts or abandoning CoT in favor of alternative methods.
Don't Wait to Reply: Towards Responsive yet Thoughtful Dialogue through Proactive Thinking
Dept. of Comp. Sci. & Tech., Institute for AI, Tsinghua University（清华大学计算机科学与技术系、人工智能研究院）；Institute for AI Industry Research (AIR), Tsinghua University（清华大学人工智能产业研究院）；Tencent（腾讯）
Don't Make the LLM Read the Graph: Make the Graph Think
Mindoverflow；University of Waterloo（多伦多大学）；Carnegie Mellon University（卡内基梅隆大学）；Foothill College（foothill学院）；Purdue University（普渡大学）；University of Wisconsin（威斯康星大学）；Apple（苹果公司）
SpatialChain: A Benchmark for Auditing Spatial Reasoning Faithfulness in VLMs
Universidade Federal de Mato Grosso (UFMT)（马托格罗索联邦大学）；Universidade Federal de Goiás (UFG)（戈亚斯联邦大学）；Advanced Knowledge Center for Immersive Technologies (AKCIT)（沉浸式技术高级知识中心）
Thinking Like a Botanist: Challenging Multimodal Language Models with Intent-Driven Chain-of-Inquiry
Department of Robotics and Mechatronics Engineering, University of Dhaka（达卡大学机器人与机电工程系）；Department of Computer Science and Engineering, University of Dhaka（达卡大学计算机科学与工程系）；Department of Agronomy, Gazipur Agricultural University（加兹ipur农业大学作物学系）；Department of Botany, University of Dhaka（达卡大学植物学系）
Beyond "I Don't Know": Evaluating LLM Self-Awareness in Discriminating Data and Model Uncertainty
Dept. of Comp. Sci. & Tech., Institute for AI, Tsinghua University, Beijing, China（计算机科学与技术系，人工智能研究院，清华大学，北京，中国）；Institute for AI Industry Research (AIR), Tsinghua University, Beijing, China（人工智能产业研究院（AIR），清华大学，北京，中国）
Measuring Chain-of-Thought Monitorability Through Faithfulness and Verbosity
ML Alignment and Theory Scholars（机器学习对齐与理论学者计划）；University of Delaware（特拉华大学）；AI Safety Argentina（阿根廷AI安全组织）；Buenos Aires AI Safety Hub（布宜诺斯艾利斯AI安全中心）；University of Buenos Aires（布宜诺斯艾利斯大学）；Center for AI Safety（AI安全中心）
Anthropic（Anthropic公司）；Machine Intelligence Research Institute（机器智能研究所）；University of California, Berkeley（加州大学伯克利分校）；University of Toronto（多伦多大学）
Our intention is to provide a definitive reference on what it would take to safely make use of generative/predictive models in the absence of a solution to the Eliciting Latent Knowledge problem. Furthermore, we believe that large language models can be understood as such predictive models of the world, and that such a conceptualization raises significant opportunities for their safe yet powerful use via carefully conditioning them to predict desirable outputs. Unfortunately, such approaches also raise a variety of potentially fatal safety problems, particularly surrounding situations where pr…
