Large language models (LLMs) such as GPT-4 sometimes appear to be creative, solving novel tasks often with a few demonstrations in the prompt. These tasks require the models to generalize on distributions different from those from training data -- which is known as out-of-distribution (OOD) generalization. Despite the tremendous success of LLMs, how they approach OOD generalization remains an open and underexplored question. We examine OOD generalization in settings where instances are generated according to hidden rules, including in-context learning with symbolic reasoning. Models are required to infer the hidden rules behind input prompts without any fine-tuning.
We empirically examined the training dynamics of Transformers on a synthetic example and conducted extensive experiments on a variety of pretrained LLMs, focusing on a type of components known as induction heads. We found that OOD generalization and composition are tied together -- models can learn rules by composing two self-attention layers, thereby achieving OOD generalization. Furthermore, a shared latent subspace in the embedding (or feature) space acts as a bridge for composition by aligning early layers and later layers, which we refer to as the common bridge representation hypothesis.
Identifying Semantic Induction Heads to Understand In-Context Learning
Shanghai Jiao Tong University（上海交通大学）；Shanghai Artificial Intelligence Laboratory（上海人工智能实验室）；Fudan University（复旦大学）；The Chinese University of Hong Kong（香港中文大学）
From Shortcut to Induction Head: How Data Diversity Shapes Algorithm Selection in Transformers
The University of Tokyo（东京大学）；RIKEN AIP（理化学研究所AIP）；Flatiron Institute（Flatiron研究所）；CNRS（法国国家科学研究中心）；Université Côte d’Azur（蔚蓝海岸大学）；New York University（纽约大学）
Can OpenAI o1 Reason Well in Ophthalmology? A 6,990-Question Head-to-Head Evaluation Study
National University of Singapore（新加坡国立大学）；Singapore Eye Research Institute（新加坡眼科研究所）；Yale University（耶鲁大学）；Cleveland Clinic（克利夫兰诊所）；Zhejiang University（浙江大学）
Listening to the Wise Few: Query-Key Alignment Unlocks Latent Correct Answers in Large Language Models
Skolkovo Institute of Science and Technology（斯科尔科沃科学技术研究院）；AI Foundation and Algorithm Lab（人工智能基金会与算法实验室）；Moscow Institute of Physics and Technology（莫斯科物理技术学院）；Artificial Intelligence Research Institute (AIRI)（人工智能研究院）；CNRS（法国国家科学研究中心）；Université Paris Cité（巴黎西岱大学）
Electrical and Computer Engineering, Carnegie Mellon University（卡内基梅隆大学电气与计算机工程系）；Language Technologies Institute, Carnegie Mellon University（卡内基梅隆大学语言技术研究所）
Most previous work on grammar induction focuses on learning phrasal or dependency structure purely from text. However, because the signal provided by text alone is limited, recently introduced visually grounded syntax models make use of multimodal information leading to improved performance in constituency grammar induction. However, as compared to dependency grammars, constituency grammars do not provide a straightforward way to incorporate visual information without enforcing language-specific heuristics. In this paper, we propose an unsupervised grammar induction model that leverages word c…
University of Illinois Urbana-Champaign（伊利诺伊大学厄巴纳-香槟分校）
Traditional event extraction methods require predefined event types and their corresponding annotations to learn event extractors. These prerequisites are often hard to be satisfied in real-world applications. This work presents a corpus-based open-domain event type induction method that automatically discovers a set of event types from a given corpus. As events of the same type could be expressed in multiple ways, we propose to represent each event type as a cluster of <predicate sense, object head> pairs. Specifically, our method (1) selects salient predicates and object heads, (2) disambigu…
TriLens: Per-Layer Logit-Lens Entropy for White-Box Hallucination Detection
Binjiang Institute of Zhejiang University（浙江大学滨海学院）；Beijing Normal-Hong Kong Baptist University（北京师范大学-香港 Baptist大学）；Zhejiang University（浙江大学）；GenTel.io；Great Bay University（Great Bay大学）
Mirror, Mirror on the Wall: Prompt Echoing in Small Instruct Language Models
Centre for Credible AI, Warsaw University of Technology（华沙理工大学可信人工智能中心）；Adam Mickiewicz University Poznan（波兹南亚当·密茨凯维奇大学）；Institute of Fundamental Technological Research, Polish Academy of Sciences（波兰科学院基础技术研究所）
