The paper introduces MIDI, a novel multilingual dataset that embeds idioms in realistic sentence and conversational contexts across diverse resource levels, revealing that idiom comprehension is signi…

The paper proposes MACAT, a Multi-Agent Culture-Aware Translation framework, to selectively translate culture-loaded words in ancient Chinese texts, achieving superior performance over existing method…

The paper demonstrates that using English prompts causes large language models to prioritize globally dominant narratives over local cultural knowledge, even when local evidence is provided.

The paper introduces IndoBias, a dual-track, culturally-grounded benchmark to evaluate biases in LLMs across Indonesian and three local languages, revealing significant differences in bias patterns ac…

The paper develops a theoretically grounded framework for evaluating multilingual LLMs in Social Sciences and Humanities, moving beyond traditional NLP benchmarks to assess interpretive validity and c…

The paper introduces CARTE, a new benchmark designed to test how well large language models understand fine-grained, regionally differentiated knowledge across the 13 metropolitan regions of France, r…

The paper investigates whether modestly sized open-source language models can grasp the semantics of rare Paired-Focus constructions, finding that understanding emerges later in training and correlate…

The paper demonstrates that the internal signals governing figurative language generation are reusable across multiple languages, showing that a steering direction learned in one language can effectiv…

The paper identifies a failure mode called spatial lexical bias in MLLMs, where adding a spatial word to options biases the model's choice, and demonstrates that this failure originates primarily from…

The paper introduces a diagnostic framework to decompose multilingual LLM performance variance, showing that language identity and model-benchmark interactions are key drivers of performance gaps.

The paper introduces CultureForest, a new benchmark for evaluating Cultural Norm Grounded Reasoning in LLMs, demonstrating that models struggle to apply their cultural knowledge effectively in realist…

The paper demonstrates that the AI-like style introduced by post-training alignment can be measured, localized, and causally removed using a novel ablation technique called PASTA.

This paper proposes a domain-specialized large language model, PoetryQwen, for precise translation and emotional understanding of classical poetry.

The paper proposes a dual-interventional framework to characterize how linguistic structures and contextual cues influence LLMs' spatial reasoning for navigation, finding that topological information…

The paper introduces RA-MoE, a novel fine-tuning framework that leverages the internal routing structure of Mixture-of-Experts (MoE) models to improve performance on multilingual downstream tasks by a…

The paper introduces a multilingual benchmark (MentalMap) to test if LLMs build internal spatial world models from text, finding a universal 'L3 reasoning cliff' suggesting that text-only working memo…

The paper investigates compositional abilities in LLMs and humans using the Personal Relation Task, finding that LLMs excel at the structured (Intensional) task while humans are better at the real-wor…

The paper proposes MIMO, a two-stage framework that improves Multilingual Information Retrieval (MLIR) by stabilizing cross-lingual alignment and enhancing retrieval discrimination using a combination…

The paper introduces ACROS, a method that induces an explicit sense representation pathway into a frozen pretrained decoder LM, enabling sense-based tasks like disambiguation and cross-lingual alignme…

The paper introduces SMILE-Next, a multimodal dataset and a novel Mixture-of-Laugh-Experts (MoLE) framework to enable large language models to robustly detect, classify, and reason about laughter in c…