In the era of large language models (LLMs), hallucination (i.e., the tendency to generate factually incorrect content) poses great challenge to trustworthy and reliable deployment of LLMs in real-world applications. To tackle the LLM hallucination, three key questions should be well studied: how to detect hallucinations (detection), why do LLMs hallucinate (source), and what can be done to mitigate them (mitigation). To address these challenges, this work presents a systematic empirical study on LLM hallucination, focused on the the three aspects of hallucination detection, source and mitigation. Specially, we construct a new hallucination benchmark HaluEval 2.0, and designs a simple yet effective detection method for LLM hallucination. Furthermore, we zoom into the different training or utilization stages of LLMs and extensively analyze the potential factors that lead to the LLM hallucination. Finally, we implement and examine a series of widely used techniques to mitigate the hallucinations in LLMs. Our work has led to several important findings to understand the hallucination origin and mitigate the hallucinations in LLMs. Our code and data can be accessed at https://github.com/RUCAIBox/HaluEval-2.0.

Large language models are known for encoding a vast amount of factual knowledge, but they often becomes outdated due to the ever-changing nature of external information. A promising solution to this challenge is the utilization of model editing methods to update the knowledge in an efficient manner. However, the majority of existing model editing techniques are limited to monolingual frameworks, thus failing to address the crucial issue of cross-lingual knowledge synchronization for multilingual models. To tackle this problem, we propose a simple yet effective method that trains multilingual patch neuron to store cross-lingual knowledge. It can be easily adapted to existing approaches to enhance their cross-lingual editing capabilities. To evaluate our method, we conduct experiments using both the XNLI dataset and a self-constructed XFEVER dataset. Experimental results demonstrate that our proposed method achieves improved performance in cross-lingual editing tasks without requiring excessive modifications to the original methodology, thereby showcasing its user-friendly characteristics. Codes will be released soon.

Defeasibility in causal reasoning implies that the causal relationship between cause and effect can be strengthened or weakened. Namely, the causal strength between cause and effect should increase or decrease with the incorporation of strengthening arguments (supporters) or weakening arguments (defeaters), respectively. However, existing works ignore defeasibility in causal reasoning and fail to evaluate existing causal strength metrics in defeasible settings. In this work, we present {\delta}-CAUSAL, the first benchmark dataset for studying defeasibility in causal reasoning. {\delta}-CAUSAL includes around 11K events spanning ten domains, featuring defeasible causality pairs, i.e., cause-effect pairs accompanied by supporters and defeaters. We further show current causal strength metrics fail to reflect the change of causal strength with the incorporation of supporters or defeaters in {\delta}-CAUSAL. To this end, we propose CESAR (Causal Embedding aSsociation with Attention Rating), a metric that measures causal strength based on token-level causal relationships. CESAR achieves a significant 69.7% relative improvement over existing metrics, increasing from 47.2% to 80.1% in capturing the causal strength change brought by supporters and defeaters. We further demonstrate even Large Language Models (LLMs) like GPT-3.5 still lag 4.5 and 10.7 points behind humans in generating supporters and defeaters, emphasizing the challenge posed by {\delta}-CAUSAL.

Medical question answer (QA) assistants respond to lay users' health-related queries by synthesizing information from multiple sources using natural language processing and related techniques. They can serve as vital tools to alleviate issues of misinformation, information overload, and complexity of medical language, thus addressing lay users' information needs while reducing the burden on healthcare professionals. QA systems, the engines of such assistants, have typically used either language models (LMs) or knowledge graphs (KG), though the approaches could be complementary. LM-based QA systems excel at understanding complex questions and providing well-formed answers, but are prone to factual mistakes. KG-based QA systems, which represent facts well, are mostly limited to answering short-answer questions with pre-created templates. While a few studies have jointly used LM and KG approaches for text-based QA, this was done to answer multiple-choice questions. Extant QA systems also have limitations in terms of automation and performance. We address these challenges by designing a novel, automated disease QA system which effectively utilizes both LM and KG techniques through a joint-reasoning approach to answer disease-related questions appropriate for lay users. Our evaluation of the system using a range of quality metrics demonstrates its efficacy over benchmark systems, including the popular ChatGPT.

Short Text Classification (STC) is crucial for processing and comprehending the brief but substantial content prevalent on contemporary digital platforms. The STC encounters difficulties in grasping semantic and syntactic intricacies, an issue that is apparent in traditional pre-trained language models. Although Graph Convolutional Networks enhance performance by integrating external knowledge bases, these methods are limited by the quality and extent of the knowledge applied. Recently, the emergence of Large Language Models (LLMs) and Chain-of-Thought (CoT) has significantly improved the performance of complex reasoning tasks. However, some studies have highlighted the limitations of their application in fundamental NLP tasks. Consequently, this study sought to employ CoT to investigate the capabilities of LLMs in STC tasks. This study introduces Quartet Logic: A Four-Step Reasoning (QLFR) framework. This framework primarily incorporates Syntactic and Semantic Enrichment CoT, effectively decomposing the STC task into four distinct steps: (i) essential concept identification, (ii) common-sense knowledge retrieval, (iii) text rewriting, and (iv) classification. This elicits the inherent knowledge and abilities of LLMs to address the challenges in STC. Surprisingly, we found that QLFR can also improve the performance of smaller models. Therefore, we developed a CoT-Driven Multi-task learning (QLFR-CML) method to facilitate the knowledge transfer from LLMs to smaller models. Extensive experimentation across six short-text benchmarks validated the efficacy of the proposed methods. Notably, QLFR achieved state-of-the-art performance on all datasets, with significant improvements, particularly on the Ohsumed and TagMyNews datasets.

Recent advances in Large Language Models (LLMs) have exhibited remarkable proficiency across various tasks. Given the potent applications of LLMs in numerous fields, there has been a surge in LLM development. In developing LLMs, a common practice involves continual pre-training on previously fine-tuned models. However, this can lead to catastrophic forgetting. In our work, we investigate the phenomenon of forgetting that occurs during continual pre-training on an existing fine-tuned LLM. We evaluate the impact of continuous pre-training on the fine-tuned LLM across various dimensions, including output format, knowledge, and reliability. Experiment results highlight the non-trivial challenge of addressing catastrophic forgetting during continual pre-training, especially the repetition issue.

The increasing use of tools and solutions based on Large Language Models (LLMs) for various tasks in the medical domain has become a prominent trend. Their use in this highly critical and sensitive domain has thus raised important questions about their robustness, especially in response to variations in input, and the reliability of the generated outputs. This study addresses these questions by constructing a textual dataset based on the ICD-10-CM code descriptions, widely used in US hospitals and containing many clinical terms, and their easily reproducible rephrasing. We then benchmarked existing embedding models, either generalist or specialized in the clinical domain, in a semantic search task where the goal was to correctly match the rephrased text to the original description. Our results showed that generalist models performed better than clinical models, suggesting that existing clinical specialized models are more sensitive to small changes in input that confuse them. The highlighted problem of specialized models may be due to the fact that they have not been trained on sufficient data, and in particular on datasets that are not diverse enough to have a reliable global language understanding, which is still necessary for accurate handling of medical documents.

This paper explores the frontiers of large language models (LLMs) in psychology applications. Psychology has undergone several theoretical changes, and the current use of Artificial Intelligence (AI) and Machine Learning, particularly LLMs, promises to open up new research directions. We provide a detailed exploration of how LLMs like ChatGPT are transforming psychological research. It discusses the impact of LLMs across various branches of psychology, including cognitive and behavioral, clinical and counseling, educational and developmental, and social and cultural psychology, highlighting their potential to simulate aspects of human cognition and behavior. The paper delves into the capabilities of these models to emulate human-like text generation, offering innovative tools for literature review, hypothesis generation, experimental design, experimental subjects, data analysis, academic writing, and peer review in psychology. While LLMs are essential in advancing research methodologies in psychology, the paper also cautions about their technical and ethical challenges. There are issues like data privacy, the ethical implications of using LLMs in psychological research, and the need for a deeper understanding of these models' limitations. Researchers should responsibly use LLMs in psychological studies, adhering to ethical standards and considering the potential consequences of deploying these technologies in sensitive areas. Overall, the article provides a comprehensive overview of the current state of LLMs in psychology, exploring potential benefits and challenges. It serves as a call to action for researchers to leverage LLMs' advantages responsibly while addressing associated risks.

With the growing use of large language models hosted on cloud platforms to offer inference services, privacy concerns are escalating, especially concerning sensitive data like investment plans and bank account details. Secure Multi-Party Computing (SMPC) emerges as a promising solution to protect the privacy of inference data and model parameters. However, the application of SMPC in Privacy-Preserving Inference (PPI) for large language models, particularly those based on the Transformer architecture, often leads to considerable slowdowns or declines in performance. This is largely due to the multitude of nonlinear operations in the Transformer architecture, which are not well-suited to SMPC and difficult to circumvent or optimize effectively. To address this concern, we introduce an advanced optimization framework called SecFormer, to achieve fast and accurate PPI for Transformer models. By implementing model design optimization, we successfully eliminate the high-cost exponential and maximum operations in PPI without sacrificing model performance. Additionally, we have developed a suite of efficient SMPC protocols that utilize segmented polynomials, Fourier series and Goldschmidt's method to handle other complex nonlinear functions within PPI, such as GeLU, LayerNorm, and Softmax. Our extensive experiments reveal that SecFormer outperforms MPCFormer in performance, showing improvements of $5.6\%$ and $24.2\%$ for BERT$_{\text{BASE}}$ and BERT$_{\text{LARGE}}$, respectively. In terms of efficiency, SecFormer is 3.56 and 3.58 times faster than Puma for BERT$_{\text{BASE}}$ and BERT$_{\text{LARGE}}$, demonstrating its effectiveness and speed.

The increased prevalence of online meetings has significantly enhanced the practicality of a model that can automatically generate the summary of a given meeting. This paper introduces a novel and effective approach to automate the generation of meeting summaries. Current approaches to this problem generate general and basic summaries, considering the meeting simply as a long dialogue. However, our novel algorithms can generate abstractive meeting summaries that are driven by the action items contained in the meeting transcript. This is done by recursively generating summaries and employing our action-item extraction algorithm for each section of the meeting in parallel. All of these sectional summaries are then combined and summarized together to create a coherent and action-item-driven summary. In addition, this paper introduces three novel methods for dividing up long transcripts into topic-based sections to improve the time efficiency of our algorithm, as well as to resolve the issue of large language models (LLMs) forgetting long-term dependencies. Our pipeline achieved a BERTScore of 64.98 across the AMI corpus, which is an approximately 4.98% increase from the current state-of-the-art result produced by a fine-tuned BART (Bidirectional and Auto-Regressive Transformers) model.