LingBench++: A Linguistically-Informed Benchmark and Reasoning Framework for Multi-Step and Cross-Cultural Inference with LLMs

We propose LingBench++, a linguistically-informed benchmark and reasoning framework designed to evaluate large language models (LLMs) on complex linguistic tasks inspired by the International Linguistics Olympiad (IOL). Unlike prior benchmarks that focus solely on final answer accuracy, LingBench++ provides structured reasoning traces, stepwise evaluation protocols, and rich typological metadata across over 90 low-resource and cross-cultural languages. We further develop a multi-agent architecture integrating grammatical knowledge retrieval, tool-augmented reasoning, and deliberate hypothesis testing. Through systematic comparisons of baseline and our proposed agentic models, we demonstrate that models equipped with external knowledge sources and iterative reasoning outperform single-pass approaches in both accuracy and interpretability. LingBench++ offers a comprehensive foundation for advancing linguistically grounded, culturally informed, and cognitively plausible reasoning in LLMs.

Continual Pre-Training is (not) What You Need in Domain Adaption

The recent advances in Legal Large Language Models (LLMs) have transformed the landscape of legal research and practice by automating tasks, enhancing research precision, and supporting complex decision-making processes. However, effectively adapting LLMs to the legal domain remains challenging due to the complexity of legal reasoning, the need for precise interpretation of specialized language, and the potential for hallucinations. This paper examines the efficacy of Domain-Adaptive Continual Pre-Training (DACP) in improving the legal reasoning capabilities of LLMs. Through a series of experiments on legal reasoning tasks within the Taiwanese legal framework, we demonstrate that while DACP enhances domain-specific knowledge, it does not uniformly improve performance across all legal tasks. We discuss the trade-offs involved in DACP, particularly its impact on model generalization and performance in prompt-based tasks, and propose directions for future research to optimize domain adaptation strategies in legal AI.

Lexical Retrieval Hypothesis in Multimodal Context

Multimodal corpora have become an essential language resource for language science and grounded natural language processing (NLP) systems due to the growing need to understand and interpret human communication across various channels. In this paper, we first present our efforts in building the first Multimodal Corpus for Languages in Taiwan (MultiMoco). Based on the corpus, we conduct a case study investigating the Lexical Retrieval Hypothesis (LRH), specifically examining whether the hand gestures co-occurring with speech constants facilitate lexical retrieval or serve other discourse functions. With detailed annotations on eight parliamentary interpellations in Taiwan Mandarin, we explore the co-occurrence between speech constants and non-verbal features (i.e., head movement, face movement, hand gesture, and function of hand gesture). Our findings suggest that while hand gestures do serve as facilitators for lexical retrieval in some cases, they also serve the purpose of information emphasis. This study highlights the potential of the MultiMoco Corpus to provide an important resource for in-depth analysis and further research in multimodal communication studies.

Exploring the Grounding Issues in Image Caption

This paper explores the grounding issue concerning multimodal semantic representation from a computational cognitive-linguistic view. Five perceptual properties of groundedness are annotated and analyzed: Affordance, Perceptual salience, Object number, Gaze cueing, and Ecological Niche Association (ENA). We annotated selected images from the Flickr30k dataset with exploratory analyses and statistical modeling of their captions. Our findings suggest that a comprehensive understanding of an object or event requires cognitive attention, semantic distinctions in linguistic expression, and multimodal construction. During this construction process, viewers integrate situated meaning and affordance into multimodal semantics, which is consolidated into image captions used in the image-text dataset incorporating visual and textual elements. Our findings suggest that situated meaning and affordance grounding are critical for grounded natural language understanding systems to generate appropriate responses and show the potential to advance the understanding of human construal in diverse situations.