Evaluating Large Language Models on the 2026 Korean CSAT Mathematics Exam: Measuring Mathematical Ability in a Zero-Data-Leakage Setting

A recent study titled 'Time-To-Inconsistency' presents a large-scale survival analysis of the robustness of Large Language Models (LLMs) against adversarial attacks, examining 36,951 dialogue turns across nine state-of-the-art models. The research reveals that abrupt semantic shifts in prompts significantly increase the likelihood of inconsistencies, while cumulative shifts may offer a protective effect, indicating adaptive conversational dynamics.

A recent study published on arXiv explores a non-linguistic model of text, focusing on a sequence of independent draws from a finite alphabet. The research reveals that word lengths follow a geometric distribution influenced by the probability of space symbols, leading to a critical word length where word types transition in frequency. This analysis has implications for understanding the structure of language models.

A recent study on Large Language Models (LLMs) highlights the challenge of context drift in multi-turn interactions, where a model's outputs may diverge from user goals over time. The research introduces a dynamical framework to analyze this drift, formalizing it through KL divergence and proposing a recurrence model to interpret its evolution. This approach aims to enhance the consistency of LLM responses across multiple conversational turns.

A recent review titled 'LLMs4All' highlights the transformative potential of Large Language Models (LLMs) across various academic disciplines, including arts, economics, and law. The paper emphasizes the capabilities of LLMs, such as ChatGPT, in generating human-like conversations and performing complex language-related tasks, suggesting significant real-world applications in fields like education and scientific discovery.

LexInstructEval has been introduced as a new benchmark and evaluation framework aimed at enhancing the ability of Large Language Models (LLMs) to follow complex lexical instructions. This framework utilizes a formal, rule-based grammar to break down intricate instructions into manageable components, facilitating a more systematic evaluation process.

A new method called generative caching has been introduced to enhance the efficiency of Large Language Models (LLMs) in handling structurally similar prompts and responses. This approach allows for the identification of reusable response patterns, achieving an impressive 83% cache hit rate while minimizing incorrect outputs in agentic workflows.

A new framework called Mujica-MyGo has been proposed to enhance multi-agent Retrieval-Augmented Generation (RAG) systems, addressing the challenges of long context lengths in large language models (LLMs). This framework aims to improve multi-turn reasoning by utilizing a divide-and-conquer approach, which helps manage the complexity of interactions with search engines during complex reasoning tasks.

The PoETa v2 benchmark has been introduced as the most extensive evaluation of Large Language Models (LLMs) for the Portuguese language, comprising over 40 tasks. This initiative aims to systematically assess more than 20 models, highlighting performance variations influenced by computational resources and language-specific adaptations. The benchmark is accessible on GitHub.