Predicting emotions elicited by news headlines can be challenging as the task is largely influenced by the varying nature of people's interpretations and backgrounds. Previous works have explored classifying discrete emotions directly from news headlines. We provide a different approach to tackling this problem by utilizing people's explanations of their emotion, written in free-text, on how they feel after reading a news headline. Using the dataset BU-NEmo+ (Gao et al., 2022), we found that for emotion classification, the free-text explanations have a strong correlation with the dominant emotion elicited by the headlines. The free-text explanations also contain more sentimental context than the news headlines alone and can serve as a better input to emotion classification models. Therefore, in this work we explored generating emotion explanations from headlines by training a sequence-to-sequence transformer model and by using pretrained large language model, ChatGPT (GPT-4). We then used the generated emotion explanations for emotion classification. In addition, we also experimented with training the pretrained T5 model for the intermediate task of explanation generation before fine-tuning it for emotion classification. Using McNemar's significance test, methods that incorporate GPT-generated free-text emotion explanations demonstrated significant improvement (P-value < 0.05) in emotion classification from headlines, compared to methods that only use headlines. This underscores the value of using intermediate free-text explanations for emotion prediction tasks with headlines.

Evaluating the bias in Large Language Models (LLMs) becomes increasingly crucial with their rapid development. However, existing evaluation methods rely on fixed-form outputs and cannot adapt to the flexible open-text generation scenarios of LLMs (e.g., sentence completion and question answering). To address this, we introduce BiasAlert, a plug-and-play tool designed to detect social bias in open-text generations of LLMs. BiasAlert integrates external human knowledge with inherent reasoning capabilities to detect bias reliably. Extensive experiments demonstrate that BiasAlert significantly outperforms existing state-of-the-art methods like GPT4-as-A-Judge in detecting bias. Furthermore, through application studies, we demonstrate the utility of BiasAlert in reliable LLM bias evaluation and bias mitigation across various scenarios. Model and code will be publicly released.

Dark patterns, design tricks used on online interfaces to manipulate users decision-making process, have raised public concerns. However, research on regulation of dark pattern remains underdeveloped and scattered, particularly regarding scholars views on the concept, regulatory paradigms, and solutions. Following PRISMA guidelines, this paper systematically reviews the formats and content of regulatory discussions on dark patterns from the interdisciplinary scholarship of Law and Human-Computer Interaction. A total of 65 studies were analysed through content and thematic analysis. This study synthesises the unique trends and characteristics of legal scholarship on dark patterns, identifying five root problems and triple layered harms. It critiques current regulations in terms of legal theories and sectoral legislations, highlighting their inadequacies in addressing dark patterns. The paper also critically examines existing proposed solutions, including paradigmatic shifts in legal doctrines, refinements to existing frameworks, technical design-embedded solutions, and accountability measures for design practices. This research critically discusses the current barriers to effective dark pattern regulations and explores promising regulatory solutions. The difficulty in identifying the normative nature of various forms of dark patterns, in identifying evident and actionable harm, and the expanding scope of dark patterns connotation inherently hinders effective regulation. However, technical design-embedded solutions, accountability frameworks, and practical design guidelines offer potential routes for more proactive regulation, while legal pluralism stands as a promising macro-level change in regulatory paradigms for dark pattern regulation.

Language models (LMs) are trained on vast amounts of text data, which may include private and copyrighted content. Data owners may request the removal of their data from a trained model due to privacy or copyright concerns. However, exactly unlearning only these datapoints (i.e., retraining with the data removed) is intractable in modern-day models. This has led to the development of many approximate unlearning algorithms. The evaluation of the efficacy of these algorithms has traditionally been narrow in scope, failing to precisely quantify the success and practicality of the algorithm from the perspectives of both the model deployers and the data owners. We address this issue by proposing MUSE, a comprehensive machine unlearning evaluation benchmark that enumerates six diverse desirable properties for unlearned models: (1) no verbatim memorization, (2) no knowledge memorization, (3) no privacy leakage, (4) utility preservation on data not intended for removal, (5) scalability with respect to the size of removal requests, and (6) sustainability over sequential unlearning requests. Using these criteria, we benchmark how effectively eight popular unlearning algorithms on 7B-parameter LMs can unlearn Harry Potter books and news articles. Our results demonstrate that most algorithms can prevent verbatim memorization and knowledge memorization to varying degrees, but only one algorithm does not lead to severe privacy leakage. Furthermore, existing algorithms fail to meet deployer's expectations because they often degrade general model utility and also cannot sustainably accommodate successive unlearning requests or large-scale content removal. Our findings identify key issues with the practicality of existing unlearning algorithms on language models, and we release our benchmark to facilitate further evaluations: muse-bench.github.io

While LLMs have demonstrated impressive performance across various domains and tasks, their security issues have become increasingly severe. Machine unlearning (MU) has emerged as a promising solution to address these issues by removing the influence of undesired data on the target model without compromising its utility in other aspects. MU typically assumes full access to the original training data to preserve utility, which is difficult to achieve in LLM unlearning. Existing LLM unlearning methods often assume access to data most affected by undesired data unlearning. However, this assumption underestimates the entanglement among various LLM capabilities and ignores data access limitations due to various issues. Moreover, these LLM unlearning methods do not sufficiently consider that unlearning requests in real-world scenarios are continuously emerging. To overcome these challenges and achieve practical LLM unlearning, we propose the O3 framework. The O3 framework includes an Out-Of-Distribution (OOD) detector to measure the similarity between input and unlearning data, and an Orthogonal low-rank adapter (LoRA) for continuously unlearning requested data. The OOD detector is trained with a novel contrastive entropy loss and utilizes a local-global layer-aggregated scoring mechanism. The orthogonal LoRA achieves parameter disentanglement among continual unlearning requests. During inference, our O3 framework can smartly decide whether and to what extent to load the unlearning LoRA based on the OOD detector's predictions. Notably, O3's effectiveness does not rely on any retained data. We conducted extensive experiments on O3 and state-of-the-art LLM unlearning methods across three tasks and seven datasets. The results indicate that O3 consistently achieves the best trade-off between unlearning effectiveness and utility preservation, especially when facing continuous unlearning requests.

OpenAI's NDAs are once again under scrutiny after whistleblowers penned a letter to the SEC alleging that employees were made to sign "illegally restrictive" agreements preventing them from speaking out on the potential harms of the company's technology. The letter, which was obtained and published online by The Washington Post, accuses OpenAI of violating SEC rules meant to protect employees' rights to report their concerns to federal authorities and prevent retaliation. It follows an official complaint that was filed with the SEC in June. In the letter, the whistleblowers ask the SEC to "take swift and aggressive steps" to enforce the rules they say OpenAI has violated. The alleged violations include making employees sign agreements "that failed to exempt disclosures of securities violations to the SEC" and requiring employees obtain consent from the company before disclosing confidential information to the authorities.

Large language models (LLMs) have exhibited their problem-solving ability in mathematical reasoning. Solving realistic optimization (OPT) problems in industrial application scenarios requires advanced and applied math ability. However, current OPT benchmarks that merely solve linear programming are far from complex realistic situations. In this work, we propose E-OPT, a benchmark for end-to-end optimization problem-solving with human-readable inputs and outputs. E-OPT contains rich optimization problems, including linear/nonlinear programming with/without table data, which can comprehensively evaluate LLMs' solving ability. In our benchmark, LLMs are required to correctly understand the problem in E-OPT and call code solver to get precise numerical answers. Furthermore, to alleviate the data scarcity for optimization problems, and to bridge the gap between open-source LLMs on a small scale (e.g., Llama-2-7b and Llama-3-8b) and closed-source LLMs (e.g., GPT-4), we further propose a novel data synthesis method namely ReSocratic. Unlike general data synthesis methods that proceed from questions to answers, ReSocratic first incrementally synthesizes optimization scenarios with mathematical formulations step by step and then back-translates the generated scenarios into questions. In such a way, we construct the ReSocratic-29k dataset from a small seed sample pool with the powerful open-source large model DeepSeek-V2. To demonstrate the effectiveness of ReSocratic, we conduct supervised fine-tuning with ReSocratic-29k on multiple open-source models. The results show that Llama3-8b is significantly improved from 13.6% to 51.7% on E-OPT, while DeepSeek-V2 reaches 61.0%, approaching 65.5% of GPT-4.

Recent advancements in Large Language Models (LLMs) have led to significant breakthroughs in various natural language processing tasks. However, generating factually consistent responses in knowledge-intensive scenarios remains a challenge due to issues such as hallucination, difficulty in acquiring long-tailed knowledge, and limited memory expansion. This paper introduces SMART, a novel multi-agent framework that leverages external knowledge to enhance the interpretability and factual consistency of LLM-generated responses. SMART comprises four specialized agents, each performing a specific sub-trajectory action to navigate complex knowledge-intensive tasks. We propose a multi-agent co-training paradigm, Long- and Short-Trajectory Learning, which ensures synergistic collaboration among agents while maintaining fine-grained execution by each agent. Extensive experiments on 5 tasks demonstrate SMART's superior performance compared to previous widely adopted methods.

Natural Language Processing (NLP) research has traditionally been predominantly focused on English, driven by the availability of resources, the size of the research community, and market demands. Recently, there has been a noticeable shift towards multilingualism in NLP, recognizing the need for inclusivity and effectiveness across diverse languages and cultures. Monolingual surveys have the potential to complement the broader trend towards multilingualism in NLP by providing foundational insights and resources necessary for effectively addressing the linguistic diversity of global communication. However, monolingual NLP surveys are extremely rare in literature. This study fills the gap by introducing a method for creating systematic and comprehensive monolingual NLP surveys. Characterized by a structured search protocol, it can be used to select publications and organize them through a taxonomy of NLP tasks. We include a classification of Language Resources (LRs), according to their availability, and datasets, according to their annotation, to highlight publicly-available and machine-actionable LRs. By applying our method, we conducted a systematic literature review of Greek NLP from 2012 to 2022, providing a comprehensive overview of the current state and challenges of Greek NLP research. We discuss the progress of Greek NLP and outline encountered Greek LRs, classified by availability and usability. As we show, our proposed method helps avoid common pitfalls, such as data leakage and contamination, and to assess language support per NLP task. We consider this systematic literature review of Greek NLP an application of our method that showcases the benefits of a monolingual NLP survey. Similar applications could be regard the myriads of languages whose progress in NLP lags behind that of well-supported languages.

Three US Senators introduced a bill that aims to rein in the rise and use of AI generated content and deepfakes by protecting the work of artists, songwriters and journalists. The Content Original Protection and Integrity from Edited and Deepfaked Media (COPIED) Act was introduced to the Senate Friday morning. The bill is a bipartisan effort authorized by Sen. Marsha Blackburn (R-Tenn.), Sen. Maria Cantwell (D-Wash.) and Sen. Martin Heinrich (D-N.M.), according to a press alert issued by Blackburn's office. The COPIED ACT would, if enacted, create transparency standards through the National Institutes of Standards and Technology (NIST) to set guidelines for "content provenance information, watermarking, and synthetic content detection," according to the press release.