Here is today's selection of cs.CL papers exploring advancements in multilingual and multimodal language technologies. The papers focus on evaluating language models' sensitivity to linguistic diversity, enhancing the detection of synthetic content in low-resource languages, and addressing language gaps in vision-language-action models.

RAVEL: Reasoning Agents for Validating and Evaluating LLM Text Synthesis

Large Language Models have evolved from single-round generators into long-horizon agents, capable of complex text synthesis scenarios. However, current evaluation frameworks lack the ability to assess the actual synthesis operations, such as outlining, drafting, and editing. Consequently, they fail to evaluate the actual and detailed capabilities of LLMs. To bridge this gap, we introduce RAVEL, an agentic framework that enables the LLM testers to autonomously plan and execute typical synthesis operations, including outlining, drafting, reviewing, and refining. Complementing this framework, we present C3EBench, a comprehensive benchmark comprising 1,258 samples derived from professional human writings. We utilize a "reverse-engineering" pipeline to isolate specific capabilities across four tasks: Cloze, Edit, Expand, and End-to-End. Through our analysis of 14 LLMs, we uncover that most LLMs struggle with tasks that demand contextual understanding under limited or under-specified instructions. By augmenting RAVEL with SOTA LLMs as operators, we find that such agentic text synthesis is dominated by the LLM's reasoning capability rather than raw generative capacity. Furthermore, we find that a strong reasoner can guide a weaker generator to yield higher-quality results, whereas the inverse does not hold. Our code and data are available at this link: https://github.com/ZhuoerFeng/RAVEL-Reasoning-Agents-Text-Eval.

A Typologically Grounded Evaluation Framework for Word Order and Morphology Sensitivity in Multilingual Masked LMs

We introduce a typology-aware diagnostic for multilingual masked language models that tests reliance on word order versus inflectional form. Using Universal Dependencies, we apply inference-time perturbations: full token scrambling, content-word scrambling with function words fixed, dependency-based head--dependent swaps, and sentence-level lemma substitution (+L), which lemmatizes both the context and the masked target label. We evaluate mBERT and XLM-R on English, Chinese, German, Spanish, and Russian. Full scrambling drives word-level reconstruction accuracy near zero in all languages; partial and head--dependent perturbations cause smaller but still large drops. +L has little effect in Chinese but substantially lowers accuracy in German/Spanish/Russian, and it does not mitigate the impact of scrambling. Top-5 word accuracy shows the same pattern: under full scrambling, the gold word rarely appears among the five highest-ranked reconstructions. We release code, sampling scripts, and balanced evaluation subsets; Turkish results under strict reconstruction are reported in the appendix.

BLUFF: Benchmarking the Detection of False and Synthetic Content across 58 Low-Resource Languages

Multilingual falsehoods threaten information integrity worldwide, yet detection benchmarks remain confined to English or a few high-resource languages, leaving low-resource linguistic communities without robust defense tools. We introduce BLUFF, a comprehensive benchmark for detecting false and synthetic content, spanning 79 languages with over 202K samples, combining human-written fact-checked content (122K+ samples across 57 languages) and LLM-generated content (79K+ samples across 71 languages). BLUFF uniquely covers both high-resource "big-head" (20) and low-resource "long-tail" (59) languages, addressing critical gaps in multilingual research on detecting false and synthetic content. Our dataset features four content types (human-written, LLM-generated, LLM-translated, and hybrid human-LLM text), bidirectional translation (English$\leftrightarrow$X), 39 textual modification techniques (36 manipulation tactics for fake news, 3 AI-editing strategies for real news), and varying edit intensities generated using 19 diverse LLMs. We present AXL-CoI (Adversarial Cross-Lingual Agentic Chainof-Interactions), a novel multi-agentic framework for controlled fake/real news generation, paired with mPURIFY, a quality filtering pipeline ensuring dataset integrity. Experiments reveal state-of-theart detectors suffer up to 25.3% F1 degradation on low-resource versus high-resource languages. BLUFF provides the research community with a multilingual benchmark, extensive linguistic-oriented benchmark evaluation, comprehensive documentation, and opensource tools to advance equitable falsehood detection. Dataset and code are available at: https://jsl5710.github.io/BLUFF/

QQ: A Toolkit for Language Identifiers and Metadata

The growing number of languages considered in multilingual NLP, including new datasets and tasks, poses challenges regarding properly and accurately reporting which languages are used and how. For example, datasets often use different language identifiers; some use BCP-47 (e.g. en_Latn), others use ISO 639-1 (en), and more linguistically oriented datasets use Glottocodes (stan1293). Mapping between identifiers is manageable for a few dozen languages, but becomes unscalable when dealing with thousands. We introduce QwanQwa, a light-weight Python toolkit for unified language metadata management. QQ integrates multiple language resources into a single interface, provides convenient normalization and mapping between language identifiers, and affords a graph-based structure that enables traversal across families, regions, writing systems, and other linguistic attributes. QQ serves both as (1) a simple "glue" library in multilingual NLP research to make working with many languages easier, and (2) as an intuitive way for exploring languages, such as finding related ones through shared scripts, regions or other metadata.

LangGap: Diagnosing and Closing the Language Gap in Vision-Language-Action Models

Vision-Language-Action (VLA) models achieve over 95% success on standard benchmarks. However, through systematic experiments, we find that current state-of-the-art VLA models largely ignore language instructions. Prior work lacks: (1) systematic semantic perturbation diagnostics, (2) a benchmark that forces language understanding by design, and (3) linguistically diverse training data. This paper constructs the LangGap benchmark, based on a four-dimensional semantic perturbation method -- varying instruction semantics while keeping the tabletop layout fixed -- revealing language understanding deficits in π0.5. Existing benchmarks like LIBERO assign only one task per layout, underutilizing available objects and target locations; LangGap fully diversifies pick-and-place tasks under identical layouts, forcing models to truly understand language. Experiments show that targeted data augmentation can partially close the language gap -- success rate improves from 0% to 90% with single-task training, and 0% to 28% with multi-task training. However, as semantic diversity of extended tasks increases, model learning capacity proves severely insufficient; even trained tasks perform poorly. This reveals a fundamental challenge for VLA models in understanding diverse language instructions -- precisely the long-term value of LangGap.