Visual token reduction lowers inference costs caused by extensive image features in large vision-language models (LVLMs). Unlike relevant studies that prune tokens in self-attention-only LVLMs, our work uniquely addresses cross-attention-based models, which achieve superior performance. We identify that the key-value (KV) cache size for image tokens in cross-attention layers significantly exceeds that of text tokens in self-attention layers, posing a major compute bottleneck. To mitigate this issue, we exploit the sparse nature in cross-attention maps to selectively prune redundant visual features. Our Trimmed Llama effectively reduces KV cache demands without requiring additional training. By benefiting from 50%-reduced visual features, our model can reduce inference latency and memory usage while achieving benchmark parity.

The release of advanced Large Language Models (LLMs) such as ChatGPT and Copilot is changing the way text is created and may influence the content that we find on the web. This study investigated whether the release of these two popular LLMs coincided with a change in writing style in headlines and links on worldwide news websites. 175 NLP features were obtained for each text in a dataset of 451 million headlines/links. An interrupted time series analysis was applied for each of the 175 NLP features to evaluate whether there were any statistically significant sustained changes after the release dates of ChatGPT and/or Copilot. There were a total of 44 features that did not appear to have any significant sustained change after the release of ChatGPT/Copilot. A total of 91 other features did show significant change with ChatGPT and/or Copilot although significance with earlier control LLM release dates (GPT-1/2/3, Gopher) removed them from consideration. This initial analysis suggests these language models may have had a limited impact on the style of individual news headlines/links, with respect to only some NLP measures.

Large language models (LLMs) have greatly improved their capability in performing NLP tasks. However, deeper semantic understanding, contextual coherence, and more subtle reasoning are still difficult to obtain. The paper discusses state-of-the-art methodologies that advance LLMs with more advanced NLU techniques, such as semantic parsing, knowledge integration, and contextual reinforcement learning. We analyze the use of structured knowledge graphs, retrieval-augmented generation (RAG), and fine-tuning strategies that match models with human-level understanding. Furthermore, we address the incorporation of transformer-based architectures, contrastive learning, and hybrid symbolic-neural methods that address problems like hallucinations, ambiguity, and inconsistency in the factual perspectives involved in performing complex NLP tasks, such as question-answering text summarization and dialogue generation. Our findings show the importance of semantic precision for enhancing AI-driven language systems and suggest future research directions to bridge the gap between statistical language models and true natural language understanding.

Large language models (LLMs) can perform complex reasoning by generating intermediate thoughts under zero-shot or few-shot settings. However, zero-shot prompting always encounters low performance, and the superior performance of few-shot prompting hinges on the manual-crafted demonstrations. In this paper, we present RoSE (Reasoning with Orchestrated Streaming Experiences), a general framework for solving reasoning tasks that can self-improve without complex external efforts. To enable RoSE, we describe an architecture that extends an LLM to store all answered questions and their thoughts in a streaming experience pool then orchestrates helpful questions from the pool to assist in answering new questions. To set up a question-aware orchestration mechanism, RoSE first calculates the similarity of each question in the pool with a new test question. Since the solution to each answered question is not always correct, RoSE will sort the questions according to their similarity with the new question, and then uniformly divide them into multiple buckets. It finally extracts one question from each bucket to make these extracted questions more diverse. To make these extracted questions help RoSE answer new questions as much as possible, we introduce two other attributes of uncertainty and complexity for each question. RoSE will preferentially select the questions with low uncertainty and high complexity from each bucket. We evaluate the versatility of RoSE in various reasoning tasks, LLMs, and CoT methods.

Generative AI has the potential to transform personalization and accessibility of education. However, it raises serious concerns about accuracy and helping students become independent critical thinkers. In this study, we designed a helpful AI "Peer" to help students correct fundamental physics misconceptions related to Newtonian mechanic concepts. In contrast to approaches that seek near-perfect accuracy to create an authoritative AI tutor or teacher, we directly inform students that this AI can answer up to 40% of questions incorrectly. In a randomized controlled trial with 165 students, those who engaged in targeted dialogue with the AI Peer achieved post-test scores that were, on average, 10.5 percentage points higher--with over 20 percentage points higher normalized gain--than a control group that discussed physics history. Qualitative feedback indicated that 91% of the treatment group's AI interactions were rated as helpful. Furthermore, by comparing student performance on pre-and post-test questions about the same concept, along with experts' annotations of the AI interactions, we find initial evidence suggesting the improvement in performance does not depend on the correctness of the AI. With further research, the AI Peer paradigm described here could open new possibilities for how we learn, adapt to, and grow with AI. Students have recently been exposed to the remarkable capabilities of Generative AI (AI) in education (AIED). For example, OpenAI's ChatGPT has been reported to successfully support teaching preparation, assessment design and grading, and student learning (Lo, 2023). Systems like ChatGPT show potential to save time and enhance teaching and learning, including critical and higher-order thinking tasks (Lo, 2023).

While football analytics has changed the way teams and analysts assess performance, there remains a communication gap between machine learning practice and how coaching staff talk about football. Coaches and practitioners require actionable insights, which are not always provided by models. To bridge this gap, we show how to build wordalizations (a novel approach that leverages large language models) for shots in football. Specifically, we first build an expected goals model using logistic regression. We then use the co-efficients of this regression model to write sentences describing how factors (such as distance, angle and defensive pressure) contribute to the model's prediction. Finally, we use large language models to give an entertaining description of the shot. We describe our approach in a model card and provide an interactive open-source application describing shots in recent tournaments. We discuss how shot wordalisations might aid communication in coaching and football commentary, and give a further example of how the same approach can be applied to other actions in football.

Extracting structured information from unstructured text is crucial for modeling real-world processes, but traditional schema mining relies on semi-structured data, limiting scalability. This paper introduces schema-miner, a novel tool that combines large language models with human feedback to automate and refine schema extraction. Through an iterative workflow, it organizes properties from text, incorporates expert input, and integrates domain-specific ontologies for semantic depth. Applied to materials science--specifically atomic layer deposition-- schema-miner demonstrates that expert-guided LLMs generate semantically rich schemas suitable for diverse real-world applications.

Mathematics learning entails mastery of both content knowledge and cognitive processing of knowing, applying, and reasoning with it. Automated math assessment primarily has focused on grading students' exhibition of content knowledge by finding textual evidence, such as specific numbers, formulas, and statements. Recent advancements in problem-solving, image recognition, and reasoning capabilities of large language models (LLMs) show promise for nuanced evaluation of students' cognitive skills. Diagnosing cognitive skills needs to infer students' thinking processes beyond textual evidence, which is an underexplored task in LLM-based automated assessment. In this work, we investigate how state-of-the-art LLMs diagnose students' cognitive skills in mathematics. We constructed MathCog, a novel benchmark dataset comprising 639 student responses to 110 expert-curated middle school math problems, each annotated with detailed teachers' diagnoses based on cognitive skill checklists. Using MathCog, we evaluated 16 closed and open LLMs of varying model sizes and vendors. Our evaluation reveals that even the state-of-the-art LLMs struggle with the task, all F1 scores below 0.5, and tend to exhibit strong false confidence for incorrect cases ($r_s=.617$). We also found that model size positively correlates with the diagnosis performance ($r_s=.771$). Finally, we discuss the implications of these findings, the overconfidence issue, and directions for improving automated cognitive skill diagnosis.

Large Language Models (LLMs) have significantly advanced Natural Language Processing (NLP), particularly in Natural Language Understanding (NLU) tasks. As we progress toward an agentic world where LLM-based agents autonomously handle specialized tasks, it becomes crucial for these models to adapt to new tasks without forgetting previously learned information - a challenge known as catastrophic forgetting. This study evaluates the continual fine-tuning of various open-source LLMs with different parameter sizes (specifically models under 10 billion parameters) on key NLU tasks from the GLUE benchmark, including SST-2, MRPC, CoLA, and MNLI. By employing prompt engineering and task-specific adjustments, we assess and compare the models' abilities to retain prior knowledge while learning new tasks. Our results indicate that models such as Phi-3.5-mini exhibit minimal forgetting while maintaining strong learning capabilities, making them well-suited for continual learning environments. Additionally, models like Orca-2-7b and Qwen2.5-7B demonstrate impressive learning abilities and overall performance after fine-tuning. This work contributes to understanding catastrophic forgetting in LLMs and highlights prompting engineering to optimize model performance for continual learning scenarios.

Large language models demonstrate remarkable reasoning capabilities but often produce unreliable or incorrect responses. Existing verification methods are typically model-specific or domain-restricted, requiring significant computational resources and lacking scalability across diverse reasoning tasks. To address these limitations, we propose VerifiAgent, a unified verification agent that integrates two levels of verification: meta-verification, which assesses completeness and consistency in model responses, and tool-based adaptive verification, where VerifiAgent autonomously selects appropriate verification tools based on the reasoning type, including mathematical, logical, or commonsense reasoning. This adaptive approach ensures both efficiency and robustness across different verification scenarios. Experimental results show that VerifiAgent outperforms baseline verification methods (e.g., deductive verifier, backward verifier) among all reasoning tasks. Additionally, it can further enhance reasoning accuracy by leveraging feedback from verification results. VerifiAgent can also be effectively applied to inference scaling, achieving better results with fewer generated samples and costs compared to existing process reward models in the mathematical reasoning domain. Code is available at https://github.com/Jiuzhouh/VerifiAgent