Is explainability a false promise? This debate has emerged from the insufficient evidence that explanations aid people in situations they are introduced for. More human-centered, application-grounded evaluations of explanations are needed to settle this. Yet, with no established guidelines for such studies in NLP, researchers accustomed to standardized proxy evaluations must discover appropriate measurements, tasks, datasets, and sensible models for human-AI teams in their studies. To help with this, we first review fitting existing metrics. We then establish requirements for datasets to be suitable for application-grounded evaluations. Among over 50 datasets available for explainability research in NLP, we find that 4 meet our criteria. By finetuning Flan-T5-3B, we demonstrate the importance of reassessing the state of the art to form and study human-AI teams. Finally, we present the exemplar studies of human-AI decision-making for one of the identified suitable tasks -- verifying the correctness of a legal claim given a contract.

Conditional Generative Adversarial Nets (CGAN) is often used to improve conditional image generation performance. However, there is little research on Representation learning with CGAN for causal inference. This paper proposes a new method for finding representation learning functions by adopting the adversarial idea. We apply the pattern of CGAN and theoretically emonstrate the feasibility of finding a suitable representation function in the context of two distributions being balanced. The theoretical result shows that when two distributions are balanced, the ideal representation function can be found and thus can be used to further research.

With the rapid advances in deep learning and smart manufacturing in Industry 4.0, there is an imperative for high-throughput, high-performance, and fully integrated visual inspection systems. Most anomaly detection approaches using defect detection datasets, such as MVTec AD, employ one-class models that require fitting separate models for each class. On the contrary, unified models eliminate the need for fitting separate models for each class and significantly reduce cost and memory requirements. Thus, in this work, we experiment with considering a unified multi-class setup. Our experimental study shows that multi-class models perform at par with one-class models for the standard MVTec AD dataset. Hence, this indicates that there may not be a need to learn separate object/class-wise models when the object classes are significantly different from each other, as is the case of the dataset considered. Furthermore, we have deployed three different unified lightweight architectures on the CPU and an edge device (NVIDIA Jetson Xavier NX). We analyze the quantized multi-class anomaly detection models in terms of latency and memory requirements for deployment on the edge device while comparing quantization-aware training (QAT) and post-training quantization (PTQ) for performance at different precision widths. In addition, we explored two different methods of calibration required in post-training scenarios and show that one of them performs notably better, highlighting its importance for unsupervised tasks. Due to quantization, the performance drop in PTQ is further compensated by QAT, which yields at par performance with the original 32-bit Floating point in two of the models considered.

Second-order information is valuable for many applications but challenging to compute. Several works focus on computing or approximating Hessian diagonals, but even this simplification introduces significant additional costs compared to computing a gradient. In the absence of efficient exact computation schemes for Hessian diagonals, we revisit an early approximation scheme proposed by Becker and LeCun (1989, BL89), which has a cost similar to gradients and appears to have been overlooked by the community. We introduce HesScale, an improvement over BL89, which adds negligible extra computation. On small networks, we find that this improvement is of higher quality than all alternatives, even those with theoretical guarantees, such as unbiasedness, while being much cheaper to compute. We use this insight in reinforcement learning problems where small networks are used and demonstrate HesScale in second-order optimization and scaling the step-size parameter. In our experiments, HesScale optimizes faster than existing methods and improves stability through step-size scaling. These findings are promising for scaling second-order methods in larger models in the future.

As people increasingly prioritize their health, the speed and breadth of health information dissemination on the internet have also grown. At the same time, the presence of false health information (health rumors) intermingled with genuine content poses a significant potential threat to public health. However, current research on Chinese health rumors still lacks a large-scale, public, and open-source dataset of health rumor information, as well as effective and reliable rumor detection methods. This paper addresses this gap by constructing a dataset containing 1.12 million health-related rumors (HealthRCN) through web scraping of common health-related questions and a series of data processing steps. HealthRCN is the largest known dataset of Chinese health information rumors to date. Based on this dataset, we propose retrieval-augmented large language models for Chinese health rumor detection and explainability (HRDE). This model leverages retrieved relevant information to accurately determine whether the input health information is a rumor and provides explanatory responses, effectively aiding users in verifying the authenticity of health information. In evaluation experiments, we compared multiple models and found that HRDE outperformed them all, including GPT-4-1106-Preview, in rumor detection accuracy and answer quality. HRDE achieved an average accuracy of 91.04% and an F1 score of 91.58%.

Generating diverse, high-quality outputs from language models is crucial for applications in education and content creation. Achieving true randomness and avoiding repetition remains a significant challenge. This study uses the Linear Congruential Generator method for systematic fact selection, combined with AI-powered content generation. We ensured unique combinations of gastrointestinal physiology and pathology facts across multiple rounds, integrating these facts into prompts for GPT-4o to create clinically relevant, vignette-style outputs. Over 14 rounds, 98 unique outputs were generated, demonstrating LCG's effectiveness in producing diverse and high-quality content. This method addresses key issues of randomness and repetition, enhancing the quality and efficiency of language model-generated content for various applications.

Large Language Models (LLMs) have gained significant attention but also raised concerns due to the risk of misuse. Jailbreak prompts, a popular type of adversarial attack towards LLMs, have appeared and constantly evolved to breach the safety protocols of LLMs. To address this issue, LLMs are regularly updated with safety patches based on reported jailbreak prompts. However, malicious users often keep their successful jailbreak prompts private to exploit LLMs. To uncover these private jailbreak prompts, extensive analysis of large-scale conversational datasets is necessary to identify prompts that still manage to bypass the system's defenses. This task is highly challenging due to the immense volume of conversation data, diverse characteristics of jailbreak prompts, and their presence in complex multi-turn conversations. To tackle these challenges, we introduce JailbreakHunter, a visual analytics approach for identifying jailbreak prompts in large-scale human-LLM conversational datasets. We have designed a workflow with three analysis levels: group-level, conversation-level, and turn-level. Group-level analysis enables users to grasp the distribution of conversations and identify suspicious conversations using multiple criteria, such as similarity with reported jailbreak prompts in previous research and attack success rates. Conversation-level analysis facilitates the understanding of the progress of conversations and helps discover jailbreak prompts within their conversation contexts. Turn-level analysis allows users to explore the semantic similarity and token overlap between a singleturn prompt and the reported jailbreak prompts, aiding in the identification of new jailbreak strategies. The effectiveness and usability of the system were verified through multiple case studies and expert interviews.

Our study explores how well the state-of-the-art Large Language Models (LLMs), like GPT-4 and Mistral, can assess the quality of scientific summaries or, more fittingly, scientific syntheses, comparing their evaluations to those of human annotators. We used a dataset of 100 research questions and their syntheses made by GPT-4 from abstracts of five related papers, checked against human quality ratings. The study evaluates both the closed-source GPT-4 and the open-source Mistral model's ability to rate these summaries and provide reasons for their judgments. Preliminary results show that LLMs can offer logical explanations that somewhat match the quality ratings, yet a deeper statistical analysis shows a weak correlation between LLM and human ratings, suggesting the potential and current limitations of LLMs in scientific synthesis evaluation.

Requiring a Large Language Model to generate intermediary reasoning steps has been shown to be an effective way of boosting performance. In fact, it has been found that instruction tuning on these intermediary reasoning steps improves model performance. In this work, we present a novel method of further improving performance by requiring models to compare multiple reasoning chains before generating a solution in a single inference step. We call this method Divergent CoT (DCoT). We find that instruction tuning on DCoT datasets boosts the performance of even smaller, and therefore more accessible, LLMs. Through a rigorous set of experiments spanning a wide range of tasks that require various reasoning types, we show that fine-tuning on DCoT consistently improves performance over the CoT baseline across model families and scales (1.3B to 70B). Through a combination of empirical and manual evaluation, we additionally show that these performance gains stem from models generating multiple divergent reasoning chains in a single inference step, indicative of the enabling of self-correction in language models. Our code and data are publicly available at https://github.com/UKPLab/arxiv2024-divergent-cot.

Making predictions in an unseen environment given data from multiple training environments is a challenging task. We approach this problem from an invariance perspective, focusing on binary classification to shed light on general nonlinear data generation mechanisms. We identify a unique form of invariance that exists solely in a binary setting that allows us to train models invariant over environments. We provide sufficient conditions for such invariance and show it is robust even when environmental conditions vary greatly. Our formulation admits a causal interpretation, allowing us to compare it with various frameworks. Finally, we propose a heuristic prediction method and conduct experiments using real and synthetic datasets.