Large language models (LLMs) can exhibit undesirable and unexpected behavior in the blink of an eye. In a recent Anthropic demo, Claude switched from coding to Googling pictures of Yellowstone, and these sudden shifts in behavior have also been observed in reasoning patterns and jailbreaks. This phenomenon is not unique to autoregressive models: in diffusion models, key features of the final output are decided in narrow ``critical windows'' of the generation process. In this work we develop a simple, unifying theory to explain this phenomenon. We show that it emerges generically as the generation process localizes to a sub-population of the distribution it models. While critical windows have been studied at length in diffusion models, existing theory heavily relies on strong distributional assumptions and the particulars of Gaussian diffusion. In contrast to existing work our theory (1) applies to autoregressive and diffusion models; (2) makes no distributional assumptions; (3) quantitatively improves previous bounds even when specialized to diffusions; and (4) requires basic tools and no stochastic calculus or statistical physics-based machinery. We also identify an intriguing connection to the all-or-nothing phenomenon from statistical inference. Finally, we validate our predictions empirically for LLMs and find that critical windows often coincide with failures in problem solving for various math and reasoning benchmarks.

This paper investigates the utilization of Large Language Models (LLMs) for solving complex linguistic puzzles, a domain requiring advanced reasoning and adept translation capabilities akin to human cognitive processes. We explore specific prompting techniques designed to enhance ability of LLMs to reason and elucidate their decision-making pathways, with a focus on Input-Output Prompting (IO), Chain-of-Thought Prompting (CoT), and Solo Performance Prompting (SPP). Utilizing datasets from the Puzzling Machine Competition and various Linguistics Olympiads, we employ a comprehensive set of metrics to assess the performance of GPT-4 0603, a prominent LLM, across these prompting methods. Our findings illuminate the potential of LLMs in linguistic reasoning and complex translation tasks, highlighting their capabilities and identifying limitations in the context of linguistic puzzles. This research contributes significantly to the broader field of Natural Language Processing (NLP) by providing insights into the optimization of LLM applications for improved reasoning and translation accuracy, thereby enriching the ongoing dialogue in NLP advancements.

The rapid advancement of large language models (LLMs) has led to a surge in both model supply and application demands. To facilitate effective matching between them, reliable, generic and efficient benchmark generators are widely needed. However, human annotators are constrained by inefficiency, and current LLM benchmark generators not only lack generalizability but also struggle with limited reliability, as they lack a comprehensive evaluation framework for validation and optimization. To fill this gap, we first propose an automated and unbiased evaluation framework, structured around four dimensions and ten criteria. Under this framework, we carefully analyze the advantages and weaknesses of directly prompting LLMs as generic benchmark generators. To enhance the reliability, we introduce a series of methods to address the identified weaknesses and integrate them as BenchMaker. Experiments across multiple LLMs and tasks confirm that BenchMaker achieves superior or comparable performance to human-annotated benchmarks on all metrics, highlighting its generalizability and reliability. More importantly, it delivers highly consistent evaluation results across 12 LLMs (0.967 Pearson correlation against MMLU-Pro), while taking only $0.005 and 0.38 minutes per sample.

The advancement of large language models (LLMs) has opened new frontiers in natural language processing, particularly in specialized domains like healthcare. In this paper, we propose the Incremental Curriculum-Based Fine-Tuning (ICFT) framework to enhance the generative capabilities of medical large language models (MLLMs). ICFT combines curriculum-based learning, dual-stage memory coordination, and parameter-efficient fine-tuning to enable a progressive transition from general linguistic knowledge to strong domain-specific expertise. Experimental results across diverse medical NLP tasks, including question answering, preference classification, and response generation, demonstrate that ICFT consistently outperforms state-of-the-art baselines, achieving improvements in both accuracy and efficiency. Further analysis reveals the framework's ability to generalize to unseen data, reduce errors, and deliver diverse, contextually relevant medical responses. These findings establish ICFT as a robust and scalable solution for adapting LLMs to the medical domain, offering practical benefits for real-world healthcare applications.

Advancements in mobility services, navigation systems, and smart transportation technologies have made it possible to collect large amounts of path data. Modeling the distribution of this path data, known as the Path Generation (PG) problem, is crucial for understanding urban mobility patterns and developing intelligent transportation systems. Recent studies have explored using diffusion models to address the PG problem due to their ability to capture multimodal distributions and support conditional generation. A recent work devises a diffusion process explicitly in graph space and achieves state-of-the-art performance. However, this method suffers a high computation cost in terms of both time and memory, which prohibits its application. In this paper, we analyze this method both theoretically and experimentally and find that the main culprit of its high computation cost is its explicit design of the diffusion process in graph space. To improve efficiency, we devise a Latent-space Path Diffusion (LPD) model, which operates in latent space instead of graph space. Our LPD significantly reduces both time and memory costs by up to 82.8% and 83.1%, respectively. Despite these reductions, our approach does not suffer from performance degradation. It outperforms the state-of-the-art method in most scenarios by 24.5%~34.0%.

Large Language Models (LLMs) are transforming how people find information, and many users turn nowadays to chatbots to obtain answers to their questions. Despite the instant access to abundant information that LLMs offer, it is still important to promote critical thinking and problem-solving skills. Automatic hint generation is a new task that aims to support humans in answering questions by themselves by creating hints that guide users toward answers without directly revealing them. In this context, hint evaluation focuses on measuring the quality of hints, helping to improve the hint generation approaches. However, resources for hint research are currently spanning different formats and datasets, while the evaluation tools are missing or incompatible, making it hard for researchers to compare and test their models. To overcome these challenges, we introduce HintEval, a Python library that makes it easy to access diverse datasets and provides multiple approaches to generate and evaluate hints. HintEval aggregates the scattered resources into a single toolkit that supports a range of research goals and enables a clear, multi-faceted, and reliable evaluation. The proposed library also includes detailed online documentation, helping users quickly explore its features and get started. By reducing barriers to entry and encouraging consistent evaluation practices, HintEval offers a major step forward for facilitating hint generation and analysis research within the NLP/IR community.

The widespread prevalence of misinformation poses significant societal concerns. Out-of-context misinformation, where authentic images are paired with false text, is particularly deceptive and easily misleads audiences. Most existing detection methods primarily evaluate image-text consistency but often lack sufficient explanations, which are essential for effectively debunking misinformation. We present a model that detects multimodal misinformation through cross-modality consistency checks, requiring minimal training time. Additionally, we propose a lightweight model that achieves competitive performance using only one-third of the parameters. We also introduce a dual-purpose zero-shot learning task for generating contextualized warnings, enabling automated debunking and enhancing user comprehension. Qualitative and human evaluations of the generated warnings highlight both the potential and limitations of our approach.

In this paper, we presents a novel method for improving text-to-image generation by combining Large Language Models (LLMs) with diffusion models, a hybrid approach aimed at achieving both higher quality and efficiency in image synthesis from text descriptions. Our approach introduces a new dynamic KL-weighting strategy to optimize the diffusion process, along with incorporating semantic understanding from pre-trained LLMs to guide the generation process. The proposed method significantly improves both the visual quality and alignment of generated images with text descriptions, addressing challenges such as computational inefficiency, instability in training, and robustness to textual variability. We evaluate our method on the COCO dataset and demonstrate its superior performance over traditional GAN-based models, both quantitatively and qualitatively. Extensive experiments, including ablation studies and human evaluations, confirm that our method outperforms existing approaches in terms of image realism, relevance to the input text, and overall aesthetic quality. Our approach also shows promise in scalability to other multimodal tasks, making it a versatile solution for a wide range of generative applications.

Node classification is a fundamental task in graph analysis, with broad applications across various fields. Recent breakthroughs in Large Language Models (LLMs) have enabled LLM-based approaches for this task. Although many studies demonstrate the impressive performance of LLM-based methods, the lack of clear design guidelines may hinder their practical application. In this work, we aim to establish such guidelines through a fair and systematic comparison of these algorithms. As a first step, we developed LLMNodeBed, a comprehensive codebase and testbed for node classification using LLMs. It includes ten datasets, eight LLM-based algorithms, and three learning paradigms, and is designed for easy extension with new methods and datasets. Subsequently, we conducted extensive experiments, training and evaluating over 2,200 models, to determine the key settings (e.g., learning paradigms and homophily) and components (e.g., model size) that affect performance. Our findings uncover eight insights, e.g., (1) LLM-based methods can significantly outperform traditional methods in a semi-supervised setting, while the advantage is marginal in a supervised setting; (2) Graph Foundation Models can beat open-source LLMs but still fall short of strong LLMs like GPT-4o in a zero-shot setting. We hope that the release of LLMNodeBed, along with our insights, will facilitate reproducible research and inspire future studies in this field. Codes and datasets are released at \href{https://llmnodebed.github.io/}{https://llmnodebed.github.io/}.

In recent years, Large Language Models (LLMs) rapidly gained popularity across all parts of society, including education. After initial skepticism and bans, many schools have chosen to embrace this new technology by integrating it into their curricula in the form of virtual tutors and teaching assistants. However, neither the companies developing this technology nor the public institutions involved in its implementation have set up a formal system to collect feedback from the stakeholders impacted by them. In this paper, we argue that understanding the perceptions of those directly affected by LLMS in the classroom, such as students and teachers, as well as those indirectly impacted, like parents and school staff, is essential for ensuring responsible use of AI in this critical domain. Our contributions are two-fold. First, we present results of a literature review focusing on the perceptions of LLM-based chatbots in education. We highlight important gaps in the literature, such as the exclusion of key educational agents (e.g., parents or school administrators) when analyzing the role of stakeholders, and the frequent omission of the learning contexts in which the AI systems are implemented. Thus, we present a taxonomy that organizes existing literature on stakeholder perceptions. Second, we propose the Contextualized Perceptions for the Adoption of Chatbots in Education (Co-PACE) framework, which can be used to systematically elicit perceptions and inform whether and how LLM-based chatbots should be designed, developed, and deployed in the classroom.