When something goes wrong, your first instinct might be to panic or shut down--but maybe it should be to ask ChatGPT for help instead. ChatGPT can provide troubleshooting help whenever you're stuck dealing with an unknown or unexpected issue. Just ask ChatGPT to provide a list of troubleshooting steps for whatever has gone wrong. In my case, I've asked what I can do about a Windows laptop that has stopped working--my short prompt of "Please list some troubleshooting steps for a Windows laptop that has stopped working" resulted in a solid list of 10 things I could try. This is equally useful for all kinds of other problems, like if your toilet randomly flushes on its own every so often, if your car starts making a weird noise, or if your wireless router is on the fritz.

This summer, EPFL and ETH Zurich will release a large language model (LLM) developed on public infrastructure. Trained on the "Alps" supercomputer at the Swiss National Supercomputing Centre (CSCS), the new LLM marks a milestone in open-source AI and multilingual excellence. Earlier this month in Geneva, around 50 leading global initiatives and organisations dedicated to open-source LLMs and trustworthy AI convened at the International Open-Source LLM Builders Summit. Hosted by the AI centres of EPFL and ETH Zurich, the event marked a significant step in building a vibrant and collaborative international ecosystem for open foundation models. Open LLMs are increasingly viewed as credible alternatives to commercial systems, most of which are developed behind closed doors in the United States or China.

Language is meaningless without context. The sentence "I'm going to war" is ominous when said by the president of the United States but reassuring when coming from a bedbug exterminator. The problem with AI chatbots is that they often strip away historical and cultural context, leading users to be confused, alarmed, or, in the worst cases, misled in harmful ways. Last week, an editor at The Atlantic reported that OpenAI's ChatGPT had praised Satan while guiding her and several colleagues through a series of ceremonies encouraging "various forms of self-mutilation." There was a bloodletting ritual called " THE RITE OF THE EDGE" as well as a days-long "deep magic" experience called "The Gate of the Devourer."

Project-specific code completion is a critical task that leverages context from a project to generate accurate code. State-of-the-art methods use retrieval-augmented generation (RAG) with large language models (LLMs) and project information for code completion. However, they often struggle to incorporate internal API information, which is crucial for accuracy, especially when APIs are not explicitly imported in the file. To address this, we propose a method to infer internal API information without relying on imports. Our method extends the representation of APIs by constructing usage examples and semantic descriptions, building a knowledge base for LLMs to generate relevant completions. We also introduce ProjBench, a benchmark that avoids leaked imports and consists of large-scale real-world projects. Experiments on ProjBench and CrossCodeEval show that our approach significantly outperforms existing methods, improving code exact match by 22.72% and identifier exact match by 18.31%. Additionally, integrating our method with existing baselines boosts code match by 47.80% and identifier match by 35.55%.

The focus on safety alignment in large language models (LLMs) has increased significantly due to their widespread adoption across different domains. The scale of LLMs play a contributing role in their success, and the growth in parameter count follows larger hidden dimensions. In this paper, we hypothesize that while the increase in dimensions has been a key advantage, it may lead to emergent problems as well. These problems emerge as the linear structures in the activation space can be exploited, in the form of activation engineering, to circumvent its safety alignment. Through detailed visualizations of linear subspaces associated with different concepts, such as safety, across various model scales, we show that the curse of high-dimensional representations uniquely impacts LLMs. Further substantiating our claim, we demonstrate that projecting the representations of the model onto a lower dimensional subspace can preserve sufficient information for alignment while avoiding those linear structures. Empirical results confirm that such dimensional reduction significantly reduces susceptibility to jailbreaking through representation engineering. Building on our empirical validations, we provide theoretical insights into these linear jailbreaking methods relative to a model's hidden dimensions. Broadly speaking, our work posits that the high dimensions of a model's internal representations can be both a blessing and a curse in safety alignment.

We present a formal operator-theoretic framework for analyzing Transformer-based language models using free probability theory. This leads to a spectral dynamic system interpretation of deep Transformers. We derive entropy-based generalization bounds under freeness assumptions and provide insight into positional encoding, spectral evolution, and representational complexity. This work offers a principled, though theoretical, perspective on structural dynamics in large language models Keywords: Transformers, Free Probability, Spectral Theory, Non-Commutative Random Variables, Language Models1. Introduction Large Language Models (LLMs) [1], particularly those based on Transformer architectures, are generative probabilistic models defined over sequences of discrete symbols.

The advent of large language models (LLMs) has led to significant achievements in various domains, including legal text processing. Leveraging LLMs for legal tasks is a natural evolution and an increasingly compelling choice. However, their capabilities are often portrayed as greater than they truly are. Despite the progress, we are still far from the ultimate goal of fully automating legal tasks using artificial intelligence (AI) and natural language processing (NLP). Moreover, legal systems are deeply domain-specific and exhibit substantial variation across different countries and languages. The need for building legal text processing applications for different natural languages is, therefore, large and urgent. However, there is a big challenge for legal NLP in low-resource languages such as Vietnamese due to the scarcity of resources and annotated data. The need for labeled legal corpora for supervised training, validation, and supervised fine-tuning is critical. In this paper, we introduce the VLQA dataset, a comprehensive and high-quality resource tailored for the Vietnamese legal domain. We also conduct a comprehensive statistical analysis of the dataset and evaluate its effectiveness through experiments with state-of-the-art models on legal information retrieval and question-answering tasks.

Recently, competition in the field of artificial intelligence (AI) has intensified among major technological companies, resulting in the continuous release of new large-language models (LLMs) that exhibit improved language understanding and context-based reasoning capabilities. It is expected that these advances will enable more efficient personalized recommendations in LLM-based recommendation systems through improved quality of training data and architectural design. However, many studies have not considered these recent developments. In this study, it was proposed to improve LLM-based recommendation systems by replacing Llama2 with Llama3 in the LlamaRec framework. To ensure a fair comparison, random seed values were set and identical input data was provided during preprocessing and training. The experimental results show average performance improvements of 38.65\%, 8.69\%, and 8.19\% for the ML-100K, Beauty, and Games datasets, respectively, thus confirming the practicality of this method. Notably, the significant improvements achieved by model replacement indicate that the recommendation quality can be improved cost-effectively without the need to make structural changes to the system. Based on these results, it is our contention that the proposed approach is a viable solution for improving the performance of current recommendation systems.

Nearly all human work is collaborative; thus, the evaluation of real-world NLP applications often requires multiple dimensions that align with diverse human perspectives. As real human evaluator resources are often scarce and costly, the emerging "LLM-as-a-judge" paradigm sheds light on a promising approach to leverage LLM agents to believably simulate human evaluators. Yet, to date, existing LLM-as-a-judge approaches face two limitations: persona descriptions of agents are often arbitrarily designed, and the frameworks are not generalizable to other tasks. To address these challenges, we propose MAJ-EVAL, a Multi-Agent-as-Judge evaluation framework that can automatically construct multiple evaluator personas with distinct dimensions from relevant text documents (e.g., research papers), instantiate LLM agents with the personas, and engage in-group debates with multi-agents to Generate multi-dimensional feedback. Our evaluation experiments in both the educational and medical domains demonstrate that MAJ-EVAL can generate evaluation results that better align with human experts' ratings compared with conventional automated evaluation metrics and existing LLM-as-a-judge methods.

Hallucinations pose critical risks for large language model (LLM)-based agents, often manifesting as hallucinative actions resulting from fabricated or misinterpreted information within the cognitive context. While recent studies have exposed such failures, existing evaluations remain fragmented and lack a principled testbed. In this paper, we present MIRAGE-Bench--Measuring Illusions in Risky AGEnt settings--the first unified benchmark for eliciting and evaluating hallucinations in interactive LLM-agent scenarios. We begin by introducing a three-part taxonomy to address agentic hallucinations: actions that are unfaithful to (i) task instructions, (ii) execution history, or (iii) environment observations. To analyze, we first elicit such failures by performing a systematic audit of existing agent benchmarks, then synthesize test cases using a snapshot strategy that isolates decision points in deterministic and reproducible manners. To evaluate hallucination behaviors, we adopt a fine-grained-level LLM-as-a-Judge paradigm with tailored risk-aware prompts, enabling scalable, high-fidelity assessment of agent actions without enumerating full action spaces. MIRAGE-Bench provides actionable insights on failure modes of LLM agents and lays the groundwork for principled progress in mitigating hallucinations in interactive environments.