Minjia Mao

Xiao Fang, Shangkun Che, Minjia Mao et al.

Large language models (LLMs) have the potential to transform our lives and work through the content they generate, known as AI-Generated Content (AIGC). To harness this transformation, we need to understand the limitations of LLMs. Here, we investigate the bias of AIGC produced by seven representative LLMs, including ChatGPT and LLaMA. We collect news articles from The New York Times and Reuters, both known for their dedication to provide unbiased news. We then apply each examined LLM to generate news content with headlines of these news articles as prompts, and evaluate the gender and racial biases of the AIGC produced by the LLM by comparing the AIGC and the original news articles. We further analyze the gender bias of each LLM under biased prompts by adding gender-biased messages to prompts constructed from these news headlines. Our study reveals that the AIGC produced by each examined LLM demonstrates substantial gender and racial biases. Moreover, the AIGC generated by each LLM exhibits notable discrimination against females and individuals of the Black race. Among the LLMs, the AIGC generated by ChatGPT demonstrates the lowest level of bias, and ChatGPT is the sole model capable of declining content generation when provided with biased prompts.

Minjia Mao, Bowen Yin, Yu Zhu et al.

Reasoning large language models (LLMs) have demonstrated superior capacities in solving complicated problems by generating long chain-of-thoughts (CoT), but such a lengthy CoT incurs high inference costs. In this study, we introduce ES-CoT, an inference-time method that shortens CoT generation by detecting answer convergence and stopping early with minimal performance loss. At the end of each reasoning step, we prompt the LLM to output its current final answer, denoted as a step answer. We then track the run length of consecutive identical step answers as a measure of answer convergence. Once the run length exhibits a sharp increase and exceeds a minimum threshold, the generation is terminated. We provide both empirical and theoretical support for this heuristic: step answers steadily converge to the final answer, and large run-length jumps reliably mark this convergence. Experiments on five reasoning datasets across three LLMs show that ES-CoT reduces the number of inference tokens by about 41\% on average while maintaining accuracy comparable to standard CoT. Further, ES-CoT integrates seamlessly with self-consistency prompting and remains robust across hyperparameter choices, highlighting it as a practical and effective approach for efficient reasoning.

Dongjun Wei, Minjia Mao, Xiao Fang et al.

The malicious usage of large language models (LLMs) has motivated the detection of LLM-generated texts. Previous work in topological data analysis shows that the persistent homology dimension (PHD) of text embeddings can serve as a more robust and promising score than other zero-shot methods. However, effectively detecting short LLM-generated texts remains a challenge. This paper presents Short-PHD, a zero-shot LLM-generated text detection method tailored for short texts. Short-PHD stabilizes the estimation of the previous PHD method for short texts by inserting off-topic content before the given input text and identifies LLM-generated text based on an established detection threshold. Experimental results on both public and generated datasets demonstrate that Short-PHD outperforms existing zero-shot methods in short LLM-generated text detection. Implementation codes are available online.

Minjia Mao, Dongjun Wei, Zeyu Chen et al.

Recent advancements in large language models (LLMs) have highlighted the risk of misusing them, raising the need for accurate detection of LLM-generated content. In response, a viable solution is to inject imperceptible identifiers into LLMs, known as watermarks. Our research extends the existing watermarking methods by proposing the novel Sampling One Then Accepting (STA-1) method. STA-1 is an unbiased watermark that preserves the original token distribution in expectation and has a lower risk of producing unsatisfactory outputs in low-entropy scenarios compared to existing unbiased watermarks. In watermark detection, STA-1 does not require prompts or a white-box LLM, provides statistical guarantees, demonstrates high efficiency in detection time, and remains robust against various watermarking attacks. Experimental results on low-entropy and high-entropy datasets demonstrate that STA-1 achieves the above properties simultaneously, making it a desirable solution for watermarking LLMs. Implementation codes for this study are available online.

Daniel Wang, Eli Brignac, Minjia Mao et al.

Large language models (LLMs) are widely applied across diverse domains, raising concerns about their limitations and potential risks. In this study, we investigate two types of bias that LLMs may display: stereotype bias and deviation bias. Stereotype bias refers to when LLMs consistently associate specific traits with a particular demographic group. Deviation bias reflects the disparity between the demographic distributions extracted from LLM-generated content and real-world demographic distributions. By asking four advanced LLMs to generate profiles of individuals, we examine the associations between each demographic group and attributes such as political affiliation, religion, and sexual orientation. Our experimental results show that all examined LLMs exhibit both significant stereotype bias and deviation bias towards multiple groups. Our findings uncover the biases that occur when LLMs infer user attributes and shed light on the potential harms of LLM-generated outputs.

Minjia Mao, Dongjun Wei, Xiao Fang et al.

The proliferation of large language models (LLMs) has significantly transformed the digital information landscape, making it increasingly challenging to distinguish between human-written and LLM-generated content. Detecting LLM-generated information is essential for preserving trust on digital platforms (e.g., social media and e-commerce sites) and preventing the spread of misinformation, a topic that has garnered significant attention in IS research. However, current detection methods, which primarily focus on identifying content generated by specific LLMs in known domains, face challenges in generalizing to new (i.e., unseen) LLMs and domains. This limitation reduces their effectiveness in real-world applications, where the number of LLMs is rapidly multiplying and content spans a vast array of domains. In response, we introduce a general LLM detector (GLD) that combines a twin memory networks design and a theory-guided detection generalization module to detect LLM-generated information across unseen LLMs and domains. Using real-world datasets, we conduct extensive empirical evaluations and case studies to demonstrate the superiority of GLD over state-of-the-art detection methods. The study has important academic and practical implications for digital platforms and LLMs.

Minjia Mao, Xiaohang Zhao, Xiao Fang

An infodemic refers to an enormous amount of true information and misinformation disseminated during a disease outbreak. Detecting misinformation at the early stage of an infodemic is key to manage it and reduce its harm to public health. An early stage infodemic is characterized by a large volume of unlabeled information concerning a disease. As a result, conventional misinformation detection methods are not suitable for this misinformation detection task because they rely on labeled information in the infodemic domain to train their models. To address the limitation of conventional methods, state-of-the-art methods learn their models using labeled information in other domains to detect misinformation in the infodemic domain. The efficacy of these methods depends on their ability to mitigate both covariate shift and concept shift between the infodemic domain and the domains from which they leverage labeled information. These methods focus on mitigating covariate shift but overlook concept shift, rendering them less effective for the task. In response, we theoretically show the necessity of tackling both covariate shift and concept shift as well as how to operationalize each of them. Built on the theoretical analysis, we develop a novel misinformation detection method that addresses both covariate shift and concept shift. Using two real-world datasets, we conduct extensive empirical evaluations to demonstrate the superior performance of our method over state-of-the-art misinformation detection methods as well as prevalent domain adaptation methods that can be tailored to solve the misinformation detection task.