Large language models (LLMs) are evolving into autonomous decision-makers, raising concerns about catastrophic risks in high-stakes scenarios, particularly in Chemical, Biological, Radiological and Nuclear (CBRN) domains. Based on the insight that such risks can originate from trade-offs between the agent's Helpful, Harmlessness and Honest (HHH) goals, we build a novel three-stage evaluation framework, which is carefully constructed to effectively and naturally expose such risks. We conduct 14,400 agentic simulations across 12 advanced LLMs, with extensive experiments and analysis. Results reveal that LLM agents can autonomously engage in catastrophic behaviors and deception, without being deliberately induced. Furthermore, stronger reasoning abilities often increase, rather than mitigate, these risks. We Figure 1: We find LLM agents can deploy catastrophic also show that these agents can violate instructions behaviors even if it has no authority and the permission and superior commands. On the whole, request is denied. It will also falsely accuse the third we empirically prove the existence of catastrophic party as a way of deception when asked by its superior.

Robust statistics aims to compute quantities to represent data where a fraction of it may be arbitrarily corrupted. The most essential statistic is the mean, and in recent years, there has been a flurry of theoretical advancement for efficiently estimating the mean in high dimensions on corrupted data. While several algorithms have been proposed that achieve near-optimal error, they all rely on large data size requirements as a function of dimension. In this paper, we perform an extensive experimentation over various mean estimation techniques where data size might not meet this requirement due to the highdimensional setting. For data with inliers generated from a Gaussian with known covariance, we find experimentally that several robust mean estimation techniques can practically improve upon the sample mean, with the quantum entropy scaling approach from Dong et.al.

Large Language Models (LLMs) have demonstrated remarkable success in various tasks such as natural language understanding, text summarization, and machine translation. However, their general-purpose nature often limits their effectiveness in domain-specific applications that require specialized knowledge, such as healthcare, chemistry, or legal analysis. To address this, researchers have explored diverse methods to enhance LLMs by integrating domain-specific knowledge. In this survey, we provide a comprehensive overview of these methods, which we categorize into four key approaches: dynamic knowledge injection, static knowledge embedding, modular adapters, and prompt optimization. Each approach offers unique mechanisms to equip LLMs with domain expertise, balancing trade-offs between flexibility, scalability, and efficiency. We discuss how these methods enable LLMs to tackle specialized tasks, compare their advantages and disadvantages, evaluate domain-specific LLMs against general LLMs, and highlight the challenges and opportunities in this emerging field. For those interested in delving deeper into this area, we also summarize the commonly used datasets and benchmarks. To keep researchers updated on the latest studies, we maintain an open-source at: https://github.com/abilliyb/Knowledge_Injection_Survey_Papers, dedicated to documenting research in the field of specialized LLM.

Deep Reinforcement Learning (RL) is remarkably effective in addressing sequential resource allocation problems in domains such as healthcare, public policy, and resource management. However, deep RL policies often lack transparency and adaptability, challenging their deployment alongside human decision-makers. In contrast, Language Agents, powered by large language models (LLMs), provide human-understandable reasoning but may struggle with effective decision making. To bridge this gap, we propose Rule-Bottleneck Reinforcement Learning (RBRL), a novel framework that jointly optimizes decision and explanations. At each step, RBRL generates candidate rules with an LLM, selects among them using an attention-based RL policy, and determines the environment action with an explanation via chain-of-thought reasoning. The RL rule selection is optimized using the environment rewards and an explainability metric judged by the LLM. Evaluations in real-world scenarios highlight RBRL's competitive performance with deep RL and efficiency gains over LLM fine-tuning. A survey further confirms the enhanced quality of its explanations.

A common strategy for fact-checking long-form content generated by Large Language Models (LLMs) is extracting simple claims that can be verified independently. Since inaccurate or incomplete claims compromise fact-checking results, ensuring claim quality is critical. However, the lack of a standardized evaluation framework impedes assessment and comparison of claim extraction methods. To address this gap, we propose a framework for evaluating claim extraction in the context of fact-checking along with automated, scalable, and replicable methods for applying this framework, including novel approaches for measuring coverage and decontextualization. We also introduce Claimify, an LLM-based claim extraction method, and demonstrate that it outperforms existing methods under our evaluation framework. A key feature of Claimify is its ability to handle ambiguity and extract claims only when there is high confidence in the correct interpretation of the source text.

The application of large language models (LLMs) in the legal domain holds significant potential for information retrieval and question answering, yet Thai legal QA systems face challenges due to a lack of standardized evaluation benchmarks and the complexity of Thai legal structures. This paper introduces NitiBench, a benchmark comprising two datasets: the NitiBench-CCL, covering general Thai financial law, and the NitiBench-Tax, which includes real-world tax law cases requiring advanced legal reasoning. We evaluate retrieval-augmented generation (RAG) and long-context LLM-based approaches to address three key research questions: the impact of domain-specific components like section-based chunking and cross-referencing, the comparative performance of different retrievers and LLMs, and the viability of long-context LLMs as an alternative to RAG. Our results show that section-based chunking significantly improves retrieval and end-to-end performance, current retrievers struggle with complex queries, and long-context LLMs still underperform RAG-based systems in Thai legal QA. To support fair evaluation, we propose tailored multi-label retrieval metrics and the use of an LLM-as-judge for coverage and contradiction detection method. These findings highlight the limitations of current Thai legal NLP solutions and provide a foundation for future research in the field. We also open-sourced our codes and dataset to available publicly.

The proliferation of social media platforms has led to an increase in the spread of hate speech, particularly targeting vulnerable communities. Unfortunately, existing methods for automatically identifying and blocking toxic language rely on pre-constructed lexicons, making them reactive rather than adaptive. As such, these approaches become less effective over time, especially when new communities are targeted with slurs not included in the original datasets. To address this issue, we present an adaptive approach that uses word embeddings to update lexicons and develop a hybrid model that adjusts to emerging slurs and new linguistic patterns. This approach can effectively detect toxic language, including intentional spelling mistakes employed by aggressors to avoid detection. Our hybrid model, which combines BERT with lexicon-based techniques, achieves an accuracy of 95% for most state-of-the-art datasets. Our work has significant implications for creating safer online environments by improving the detection of toxic content and proactively updating the lexicon. Content Warning: This paper contains examples of hate speech that may be triggering.

The use of ChatGPT to analyze and classify evidence in criminal proceedings has been a topic of ongoing discussion. However, to the best of our knowledge, this issue has not been studied in the context of the Polish language. This study addresses this research gap by evaluating the effectiveness of ChatGPT in classifying legal cases under the Polish Penal Code. The results show excellent binary classification accuracy, with all positive and negative cases correctly categorized. In addition, a qualitative evaluation confirms that the legal basis provided for each case, along with the relevant legal content, was appropriate. The results obtained suggest that ChatGPT can effectively analyze and classify evidence while applying the appropriate legal rules. In conclusion, ChatGPT has the potential to assist interested parties in the analysis of evidence and serve as a valuable legal resource for individuals with less experience or knowledge in this area.

OpenAI on Friday rejected a 97.4bn bid from a consortium led by billionaire Elon Musk for the ChatGPT maker, saying the startup is not for sale. The unsolicited approach is Musk's latest attempt to block the startup he co-founded with CEO Sam Altman – but later left – from becoming a for-profit firm, as it looks to secure more capital and stay ahead in the AI race. "OpenAI is not for sale, and the board has unanimously rejected Mr Musk's latest attempt to disrupt his competition. Any potential reorganization of OpenAI will strengthen our nonprofit and its mission to ensure AGI benefits all of humanity," OpenAI said on X, quoting its chair Bret Taylor, on behalf of its board. On Tuesday, Altman told news website Axios that OpenAI was not for sale.

Kurt "The Cyberguy" Knutsson explains how facial recognition technology can help you find your perfect match. From a poisonous date to finding love with a serial killer, these six chilling cases show how unsuspecting dating app users on the quest for romance led them into the clutches of danger. Dating apps – from Tinder to Grindr – are the modern way for people to connect with potential partners from the comfort of their own space. Brace yourself for stories that blur the line between love and terror. Here is Fox News Digital's list of some recent cases where love went wrong.