This paper introduces v0.5 of the AI Safety Benchmark, which has been created by the MLCommons AI Safety Working Group. The AI Safety Benchmark has been designed to assess the safety risks of AI systems that use chat-tuned language models. We introduce a principled approach to specifying and constructing the benchmark, which for v0.5 covers only a single use case (an adult chatting to a general-purpose assistant in English), and a limited set of personas (i.e., typical users, malicious users, and vulnerable users). We created a new taxonomy of 13 hazard categories, of which 7 have tests in the v0.5 benchmark. We plan to release version 1.0 of the AI Safety Benchmark by the end of 2024. The v1.0 benchmark will provide meaningful insights into the safety of AI systems. However, the v0.5 benchmark should not be used to assess the safety of AI systems. We have sought to fully document the limitations, flaws, and challenges of v0.5. This release of v0.5 of the AI Safety Benchmark includes (1) a principled approach to specifying and constructing the benchmark, which comprises use cases, types of systems under test (SUTs), language and context, personas, tests, and test items; (2) a taxonomy of 13 hazard categories with definitions and subcategories; (3) tests for seven of the hazard categories, each comprising a unique set of test items, i.e., prompts. There are 43,090 test items in total, which we created with templates; (4) a grading system for AI systems against the benchmark; (5) an openly available platform, and downloadable tool, called ModelBench that can be used to evaluate the safety of AI systems on the benchmark; (6) an example evaluation report which benchmarks the performance of over a dozen openly available chat-tuned language models; (7) a test specification for the benchmark.
The ability to accurately identify authorship is crucial for verifying content authenticity and mitigating misinformation. Large Language Models (LLMs) have demonstrated exceptional capacity for reasoning and problem-solving. However, their potential in authorship analysis, encompassing authorship verification and attribution, remains underexplored. This paper conducts a comprehensive evaluation of LLMs in these critical tasks. Traditional studies have depended on hand-crafted stylistic features, whereas state-of-the-art approaches leverage text embeddings from pre-trained language models. These methods, which typically require fine-tuning on labeled data, often suffer from performance degradation in cross-domain applications and provide limited explainability. This work seeks to address three research questions: (1) Can LLMs perform zero-shot, end-to-end authorship verification effectively? (2) Are LLMs capable of accurately attributing authorship among multiple candidates authors (e.g., 10 and 20)? (3) How can LLMs provide explainability in authorship analysis, particularly through the role of linguistic features? Moreover, we investigate the integration of explicit linguistic features to guide LLMs in their reasoning processes. Our extensive assessment demonstrates LLMs' proficiency in both tasks without the need for domain-specific fine-tuning, providing insights into their decision-making via a detailed analysis of linguistic features. This establishes a new benchmark for future research on LLM-based authorship analysis. The code and data are available at https://github.com/baixianghuang/authorship-llm.
Large Language Model (LLM) agents have been increasingly adopted as simulation tools to model humans in applications such as social science. However, one fundamental question remains: can LLM agents really simulate human behaviors? In this paper, we focus on one of the most critical behaviors in human interactions, trust, and aim to investigate whether or not LLM agents can simulate human trust behaviors. We first find that LLM agents generally exhibit trust behaviors, referred to as agent trust, under the framework of Trust Games, which are widely recognized in behavioral economics. Then, we discover that LLM agents can have high behavioral alignment with humans regarding trust behaviors, indicating the feasibility to simulate human trust behaviors with LLM agents. In addition, we probe into the biases in agent trust and the differences in agent trust towards agents and humans. We also explore the intrinsic properties of agent trust under conditions including advanced reasoning strategies and external manipulations. We further offer important implications for various scenarios where trust is paramount. Our study represents a significant step in understanding the behaviors of LLM agents and the LLM-human analogy.
The advent of Large Language Models (LLMs) has made a transformative impact. However, the potential that LLMs such as ChatGPT can be exploited to generate misinformation has posed a serious concern to online safety and public trust. A fundamental research question is: will LLM-generated misinformation cause more harm than human-written misinformation? We propose to tackle this question from the perspective of detection difficulty. We first build a taxonomy of LLM-generated misinformation. Then we categorize and validate the potential real-world methods for generating misinformation with LLMs. Then, through extensive empirical investigation, we discover that LLM-generated misinformation can be harder to detect for humans and detectors compared to human-written misinformation with the same semantics, which suggests it can have more deceptive styles and potentially cause more harm. We also discuss the implications of our discovery on combating misinformation in the age of LLMs and the countermeasures.
Graph anomaly detection has long been an important problem in various domains pertaining to information security such as financial fraud, social spam, network intrusion, etc. The majority of existing methods are performed in an unsupervised manner, as labeled anomalies in a large scale are often too expensive to acquire. However, the identified anomalies may turn out to be data noises or uninteresting data instances due to the lack of prior knowledge on the anomalies. In realistic scenarios, it is often feasible to obtain limited labeled anomalies, which have great potential to advance graph anomaly detection. However, the work exploring limited labeled anomalies and a large amount of unlabeled nodes in graphs to detect anomalies is rather limited. Therefore, in this paper, we study a novel problem of few-shot graph anomaly detection. We propose a new framework MetaGAD to learn to meta-transfer the knowledge between unlabeled and labeled nodes for graph anomaly detection. Experimental results on six real-world datasets with synthetic anomalies and "organic" anomalies (available in the dataset) demonstrate the effectiveness of the proposed approach in detecting anomalies with limited labeled anomalies.
Social media has been one of the main information consumption sources for the public, allowing people to seek and spread information more quickly and easily. However, the rise of various social media platforms also enables the proliferation of online misinformation. In particular, misinformation in the health domain has significant impacts on our society such as the COVID-19 infodemic. Therefore, health misinformation in social media has become an emerging research direction that attracts increasing attention from researchers of different disciplines. Compared to misinformation in other domains, the key differences of health misinformation include the potential of causing actual harm to humans' bodies and even lives, the hardness to identify for normal people, and the deep connection with medical science. In addition, health misinformation on social media has distinct characteristics from conventional channels such as television on multiple dimensions including the generation, dissemination, and consumption paradigms. Because of the uniqueness and importance of combating health misinformation in social media, we conduct this survey to further facilitate interdisciplinary research on this problem. In this survey, we present a comprehensive review of existing research about online health misinformation in different disciplines. Furthermore, we also systematically organize the related literature from three perspectives: characterization, detection, and intervention. Lastly, we conduct a deep discussion on the pressing open issues of combating health misinformation in social media and provide future directions for multidisciplinary researchers.
Machine learning models have demonstrated promising performance in many areas. However, the concerns that they can be biased against specific groups hinder their adoption in high-stake applications. Thus it is essential to ensure fairness in machine learning models. Most of the previous efforts require access to sensitive attributes for mitigating bias. Nonetheless, it is often infeasible to obtain large scale of data with sensitive attributes due to people's increasing awareness of privacy and the legal compliance. Therefore, an important research question is how to make fair predictions under privacy? In this paper, we study a novel problem on fair classification in a semi-private setting, where most of the sensitive attributes are private and only a small amount of clean sensitive attributes are available. To this end, we propose a novel framework FairSP that can first learn to correct the noisy sensitive attributes under privacy guarantee via exploiting the limited clean sensitive attributes. Then, it jointly models the corrected and clean data in an adversarial way for debiasing and prediction. Theoretical analysis shows that the proposed model can ensure fairness when most of the sensitive attributes are private. Experimental results on real-world datasets demonstrate the effectiveness of the proposed model for making fair predictions under privacy and maintaining high accuracy.
Graph outlier detection is an emerging but crucial machine learning task with numerous applications. Despite the proliferation of algorithms developed in recent years, the lack of a standard and unified setting for performance evaluation limits their advancement and usage in real-world applications. To tap the gap, we present, (to our best knowledge) the first comprehensive unsupervised node outlier detection benchmark for graphs called UNOD, with the following highlights: (1) evaluating fourteen methods with backbone spanning from classical matrix factorization to the latest graph neural networks; (2) benchmarking the method performance with different types of injected outliers and organic outliers on real-world datasets; (3) comparing the efficiency and scalability of the algorithms by runtime and GPU memory usage on synthetic graphs at different scales. Based on the analyses of extensive experimental results, we discuss the pros and cons of current UNOD methods, and point out multiple crucial and promising future research directions.
Modern machine learning (ML) models are becoming increasingly popular and are widely used in decision-making systems. However, studies have shown critical issues of ML discrimination and unfairness, which hinder their adoption on high-stake applications. Recent research on fair classifiers has drawn significant attention to develop effective algorithms to achieve fairness and good classification performance. Despite the great success of these fairness-aware machine learning models, most of the existing models require sensitive attributes to preprocess the data, regularize the model learning or postprocess the prediction to have fair predictions. However, sensitive attributes are often incomplete or even unavailable due to privacy, legal or regulation restrictions. Though we lack the sensitive attribute for training a fair model in the target domain, there might exist a similar domain that has sensitive attributes. Thus, it is important to exploit auxiliary information from the similar domain to help improve fair classification in the target domain. Therefore, in this paper, we study a novel problem of exploring domain adaptation for fair classification. We propose a new framework that can simultaneously estimate the sensitive attributes while learning a fair classifier in the target domain. Extensive experiments on real-world datasets illustrate the effectiveness of the proposed model for fair classification, even when no sensitive attributes are available in the target domain.
Recent advances on large pre-trained language models (PLMs) lead impressive gains on natural language understanding (NLU) tasks with task-specific fine-tuning. However, direct fine-tuning PLMs heavily relies on large amount of labeled instances, which are expensive and time-consuming to obtain. Prompt-based tuning on PLMs has proven valuable for few shot tasks. Existing works studying prompt-based tuning for few-shot NLU mainly focus on deriving proper label words with a verbalizer or generating prompt templates for eliciting semantics from PLMs. In addition, conventional data augmentation methods have also been verified useful for few-shot tasks. However, there currently are few data augmentation methods designed for the prompt-based tuning paradigm. Therefore, we study a new problem of data augmentation for prompt-based few shot learners. Since label semantics are helpful in prompt-based tuning, we propose a novel label-guided data augmentation method PromptDA which exploits the enriched label semantic information for data augmentation. Experimental results on several few shot text classification tasks show that our proposed framework achieves superior performance by effectively leveraging label semantics and data augmentation in language understanding.