Shortened version with additional results from CODEX and GPT-3. The authors contributed equally. Work done when Aman Madaan was a student researcher at Google Research, Brain Team
The past decade has witnessed dramatic gains in natural language processing and an unprecedented scaling of large language models. These developments have been accelerated by the advent of few-shot techniques such as chain of thought (CoT) prompting. Specifically, CoT pushes the performance of large language models in a few-shot setup by augmenting the prompts with intermediate steps. Despite impressive results across various tasks, the reasons behind their success have not been explored. This work uses counterfactual prompting to develop a deeper understanding of CoT-based few-shot prompting mechanisms in large language models. We first systematically identify and define the key components of a prompt: symbols, patterns, and text. Then, we devise and conduct an exhaustive set of experiments across four different tasks, by querying the model with counterfactual prompts where only one of these components is altered. Our experiments across three models (PaLM, GPT-3, and CODEX) reveal several surprising findings and brings into question the conventional wisdom around few-shot prompting. First, the presence of factual patterns in a prompt is practically immaterial to the success of CoT. Second, our results conclude that the primary role of intermediate steps may not be to facilitate learning how to solve a task. The intermediate steps are rather a beacon for the model to realize what symbols to replicate in the output to form a factual answer. Further, text imbues patterns with commonsense knowledge and meaning. Our empirical and qualitative analysis reveals that a symbiotic relationship between text and patterns explains the success of few-shot prompting: text helps extract commonsense from the question to help patterns, and patterns enforce task understanding and direct text generation.
Georgia Institute of Technology（佐治亚理工学院）；Facebook AI Research（Facebook 人工智能研究院）
Theory of Mind is the ability to attribute mental states (beliefs, intents, knowledge, perspectives, etc.) to others and recognize that these mental states may differ from one's own. Theory of Mind is critical to effective communication and to teams demonstrating higher collective performance. To effectively leverage the progress in Artificial Intelligence (AI) to make our lives more productive, it is important for humans and AI to work well together in a team. Traditionally, there has been much emphasis on research to make AI more accurate, and (to a lesser extent) on having it better underst…
National Research University Higher School of Economics（国立研究大学高等经济学院）；Qatar Computing Research Institute, HBKU（卡塔尔计算研究所，哈马德·本·哈利法大学）
We describe our system for SemEval-2020 Task 11 on Detection of Propaganda Techniques in News Articles. We developed ensemble models using RoBERTa-based neural architectures, additional CRF layers, transfer learning between the two subtasks, and advanced post-processing to handle the multi-label nature of the task, the consistency between nested spans, repetitions, and labels from similar spans in training. We achieved sizable improvements over baseline fine-tuned RoBERTa models, and the official evaluation ranked our system 3rd (almost tied with the 2nd) out of 36 teams on the span identifica…
Large language models have been shown to struggle with multi-step reasoning, and do not retain previous reasoning steps for future use. We propose a simple method for solving both of these problems by allowing the model to take Self-Notes. Unlike recent chain-of-thought or scratchpad approaches, the model can deviate from the input context at any time to explicitly think and write down its thoughts. This allows the model to perform reasoning on the fly as it reads the context and even integrate previous reasoning steps, thus enhancing its memory with useful information and enabling multi-step …
Visual Thoughts: A Unified Perspective of Understanding Multimodal Chain-of-Thought
School of Computer Science and Engineering, Central South University（中南大学计算机科学与工程学院）；Research Center for Social Computing and Interactive Robotics, Harbin Institute of Technology（哈尔滨工业大学社会计算与交互机器人研究中心）；Institute of Computing and Intelligence, Harbin Institute of Technology, Shenzhen（哈尔滨工业大学深圳研究院计算与智能研究所）；Text Computing and Cognitive Intelligence Ministry of Education Engineering Research Center, Guizhou University（贵州大学文字计算与认知智能教育部工程研究中心）；Chinese University of Hong Kong（香港中文大学）；Shanghai AI Laboratory（上海人工智能实验室）；National University of Singapore（新加坡国立大学）；Peking University（北京大学）；ByteDance Seed (China)（字节跳动种子（中国））
Can Reasoning Path still be Effective as Input? Bridging Post-Reasoning to Chain-of-Thought Compression
Faculty of Electronic and Information Engineering, Xi’an Jiaotong University（西安交通大学电子与信息工程学院）；Queen Mary University of London（伦敦大学玛丽女王学院）；City University of Hong Kong（香港城市大学）
M$^3$CoT: A Novel Benchmark for Multi-Domain Multi-step Multi-modal Chain-of-Thought
Research Center for Social Computing and Information Retrieval（社会计算与信息检索研究中心）；Harbin Institute of Technology（哈尔滨工业大学）；School of Computer Science and Engineering, Central South University（中南大学计算机科学与工程学院）；Shanghai AI Laboratory（上海人工智能实验室）
Are language models aware of the road not taken? Token-level uncertainty and hidden state dynamics
Department of Linguistics, Stanford University（斯坦福大学语言学系）；Department of Psychology, Harvard University（哈佛大学心理学系）；Center for Brain Science, Harvard University（哈佛大学脑科学中心）；Physics of Intelligence Group, NTT Research（NTT研究物理智能小组）
Tsinghua University（清华大学）；Institute for Artificial Intelligence（人工智能研究院）；State Key Lab of Intelligent Technology and Systems（智能技术与系统国家重点实验室）；Beijing National Research Center for Information Science and Technology（北京信息科学与技术国家研究中心）
Given that rich information is hidden behind ubiquitous numbers in text, numerical reasoning over text should be an essential skill of AI systems. To derive precise equations to solve numerical reasoning problems, previous work focused on modeling the structures of equations, and has proposed various structured decoders. Though structure modeling proves to be effective, these structured decoders construct a single equation in a pre-defined autoregressive order, potentially placing an unnecessary restriction on how a model should grasp the reasoning process. Intuitively, humans may have numerou…
