In-Context Learning (ICL) enables large language models (LLMs) to achieve rapid task adaptation by learning from demonstrations. With the increase in available context length of LLMs, recent experiments have shown that the performance of ICL does not necessarily scale well in many-shot (demonstration) settings. We theoretically and experimentally confirm that the reason lies in more demonstrations dispersing the model attention from the query, hindering its understanding of key content. Inspired by how humans learn from examples, we propose a training-free method FocusICL, which conducts triviality filtering to avoid attention being diverted by unimportant contents at token-level and operates hierarchical attention to further ensure sufficient attention towards current query at demonstration-level. We also design an efficient hyperparameter searching strategy for FocusICL based on model perplexity of demonstrations. Comprehensive experiments validate that FocusICL achieves an average performance improvement of 5.2% over vanilla ICL and scales well with many-shot demonstrations.

Medical abstractive summarization faces the challenge of balancing faithfulness and informativeness. Current methods often sacrifice key information for faithfulness or introduce confabulations when prioritizing informativeness. While recent advancements in techniques like in-context learning (ICL) and fine-tuning have improved medical summarization, they often overlook crucial aspects such as faithfulness and informativeness without considering advanced methods like model reasoning and self-improvement. Moreover, the field lacks a unified benchmark, hindering systematic evaluation due to varied metrics and datasets. This paper addresses these gaps by presenting a comprehensive benchmark of six advanced abstractive summarization methods across three diverse datasets using five standardized metrics. Building on these findings, we propose uMedSum, a modular hybrid summarization framework that introduces novel approaches for sequential confabulation removal followed by key missing information addition, ensuring both faithfulness and informativeness. Our work improves upon previous GPT-4-based state-of-the-art (SOTA) medical summarization methods, significantly outperforming them in both quantitative metrics and qualitative domain expert evaluations. Notably, we achieve an average relative performance improvement of 11.8% in reference-free metrics over the previous SOTA. Doctors prefer uMedSum's summaries 6 times more than previous SOTA in difficult cases where there are chances of confabulations or missing information. These results highlight uMedSum's effectiveness and generalizability across various datasets and metrics, marking a significant advancement in medical summarization.

Analog and mixed-signal (A/MS) integrated circuits (ICs) are crucial in modern electronics, playing key roles in signal processing, amplification, sensing, and power management. Many IC companies outsource manufacturing to third-party foundries, creating security risks such as stealthy analog Trojans. Traditional detection methods, including embedding circuit watermarks or conducting hardware-based monitoring, often impose significant area and power overheads, and may not effectively identify all types of Trojans. To address these shortcomings, we propose SPICED, a Large Language Model (LLM)-based framework that operates within the software domain, eliminating the need for hardware modifications for Trojan detection and localization. This is the first work using LLM-aided techniques for detecting and localizing syntactical bugs and analog Trojans in circuit netlists, requiring no explicit training and incurring zero area overhead. Our framework employs chain-of-thought reasoning and few-shot examples to teach anomaly detection rules to LLMs. With the proposed method, we achieve an average Trojan coverage of 93.32% and an average true positive rate of 93.4% in identifying Trojan-impacted nodes for the evaluated analog benchmark circuits. These experimental results validate the effectiveness of LLMs in detecting and locating both syntactical bugs and Trojans within analog netlists.

Large Language Models (LLMs) excel at generating human-like dialogues and comprehending text. However, understanding the subtleties of complex exchanges in language remains a challenge. We propose a bootstrapping framework that leverages self-generated feedback to enhance LLM reasoning capabilities for lie detection. The framework consists of three stages: suggestion, feedback collection, and modification. In the suggestion stage, a cost-effective language model generates initial predictions based on game state and dialogue. The feedback-collection stage involves a language model providing feedback on these predictions. In the modification stage, a more advanced language model refines the initial predictions using the auto-generated feedback. We investigate the application of the proposed framework for detecting betrayal and deception in Diplomacy games, and compare it with feedback from professional human players. The LLM-generated feedback exhibits superior quality and significantly enhances the performance of the model. Our approach achieves a 39% improvement over the zero-shot baseline in lying-F1 without the need for any training data, rivaling state-of-the-art supervised learning results.

Existing datasets for tabular question answering typically focus exclusively on text within cells. However, real-world data is inherently multimodal, often blending images such as symbols, faces, icons, patterns, and charts with textual content in tables. With the evolution of AI models capable of multimodal reasoning, it is pertinent to assess their efficacy in handling such structured data. This study investigates whether current AI models can perform knowledge-aware reasoning on multimodal structured data. We explore their ability to reason on tables that integrate both images and text, introducing MMTabQA, a new dataset designed for this purpose. Our experiments highlight substantial challenges for current AI models in effectively integrating and interpreting multiple text and image inputs, understanding visual context, and comparing visual content across images. These findings establish our dataset as a robust benchmark for advancing AI's comprehension and capabilities in analyzing multimodal structured data.

Parents are starting to fret about lunch packs, school uniforms and schoolbooks. School leavers who have university places are wondering what freshers' week will be like. And some university professors, especially in the humanities, will be apprehensively pondering how to deal with students who are already more adept users of large language models (LLMs) than they are. They're right to be concerned. As Ian Bogost, a professor of film and media and computer science at Washington University in St Louis, puts it: "If the first year of AI college ended in a feeling of dismay, the situation has now devolved into absurdism. Teachers struggle to continue teaching even as they wonder whether they are grading students or computers; in the meantime, an endless AI cheating and detection arms race plays out in the background."

To evaluate Large Language Models (LLMs) for question answering (QA), traditional methods typically focus on directly assessing the immediate responses generated by the models based on the given question and context. In the common use case of humans seeking AI assistant's help in finding information, these non-interactive evaluations do not account for the dynamic nature of human-model conversations, and interaction-aware evaluations have shown that accurate QA models are preferred by humans (Lee et al., 2023). Recent works in human-computer interaction (HCI) have employed human evaluators to conduct interactions and evaluations, but they are often prohibitively expensive and time-consuming to scale. In this work, we introduce an automatic evaluation framework IQA-EVAL to Interactive Question Answering Evaluation. More specifically, we introduce LLM-based Evaluation Agent (LEA) that can: (1) simulate human behaviors to generate interactions with IQA models; (2) automatically evaluate the generated interactions. Moreover, we propose assigning personas to LEAs to better simulate groups of real human evaluators. We show that: (1) our evaluation framework with GPT-4 (or Claude) as the backbone model achieves a high correlation with human evaluations on the IQA task; (2) assigning personas to LEA to better represent the crowd further significantly improves correlations. Finally, we use our automatic metric to evaluate five recent representative LLMs with over 1000 questions from complex and ambiguous question answering tasks, which comes with a substantial cost of $5k if evaluated by humans.

The stock price prediction task holds a significant role in the financial domain and has been studied for a long time. Recently, large language models (LLMs) have brought new ways to improve these predictions. While recent financial large language models (FinLLMs) have shown considerable progress in financial NLP tasks compared to smaller pre-trained language models (PLMs), challenges persist in stock price forecasting. Firstly, effectively integrating the modalities of time series data and natural language to fully leverage these capabilities remains complex. Secondly, FinLLMs focus more on analysis and interpretability, which can overlook the essential features of time series data. Moreover, due to the abundance of false and redundant information in financial markets, models often produce less accurate predictions when faced with such input data. In this paper, we introduce StockTime, a novel LLM-based architecture designed specifically for stock price data. Unlike recent FinLLMs, StockTime is specifically designed for stock price time series data. It leverages the natural ability of LLMs to predict the next token by treating stock prices as consecutive tokens, extracting textual information such as stock correlations, statistical trends and timestamps directly from these stock prices. StockTime then integrates both textual and time series data into the embedding space. By fusing this multimodal data, StockTime effectively predicts stock prices across arbitrary look-back periods. Our experiments demonstrate that StockTime outperforms recent LLMs, as it gives more accurate predictions while reducing memory usage and runtime costs.

To enhance the reasoning capabilities of large language models (LLMs), self-consistency has gained significant popularity by combining multiple sampling with majority voting. However, the state-of-the-art self-consistency approaches consume substantial computational resources and lead to significant additional time costs due to the multiple sampling. This prevents its full potential from being realized in scenarios where computational resources are critical. To improve the inference efficiency, this paper introduces \textit{path-consistency}, a method that leverages the confidence of answers generated in earlier branches to identify the prefix of the most promising path. By dynamically guiding the generation of subsequent branches based on this prefix, the \textit{path-consistency} mitigates both the errors and redundancies from random or less useful sampling in self-consistency. As a result, it can significantly accelerate the inference process by reducing the number of tokens generated. Our extensive empirical evaluation shows that the \textit{path-consistency} achieves significant acceleration in inference latency ranging from $7.8\%$ to $40.5\%$, while maintaining or even improving task accuracy across different datasets, including mathematical reasoning, common sense reasoning, symbolic reasoning, and code generation.

Nowadays, Large Language Models (LLMs) have demonstrated exceptional performance across various downstream tasks. However, it is challenging for users to discern whether the responses are generated with certainty or are fabricated to meet user expectations. Estimating the uncertainty of LLMs is particularly challenging due to their vast scale and the lack of white-box access. In this work, we propose a novel Uncertainty Tripartite Testing Paradigm (Unc-TTP) to classify LLM uncertainty, via evaluating the consistency of LLM outputs when incorporating label interference into the sampling-based approach. Based on Unc-TTP outputs, we aggregate instances into certain and uncertain categories. Further, we conduct a detailed analysis of the uncertainty properties of LLMs and show Unc-TTP's superiority over the existing sampling-based methods. In addition, we leverage the obtained uncertainty information to guide in-context example selection, demonstrating that Unc-TTP obviously outperforms retrieval-based and sampling-based approaches in selecting more informative examples. Our work paves a new way to classify the uncertainty of both open- and closed-source LLMs, and introduces a practical approach to exploit this uncertainty to improve LLMs performance.