Multilingual

Large Language Models (LLMs) have made remarkable progress in the processing and modeling of many languages. Yet, unlike human multilinguals, they exhibit surprisingly limited cross-lingual knowledge transfer. While this limitation is well documented, its origins during multilingual training remain unclear. We pretrain 360M- and 7B-parameter LLMs and show that poor cross-lingual knowledge generalization emerges during pretraining and persists under standard interventions. To isolate its cause, we employ a controlled bilingual pretraining setting using two copies of the same language, sharing identical text and token segmentation, but mapped to disjoint token spaces. We find that disjoint tokens alone are enough to induce knowledge compartmentalization, even between identical copies of the same language, establishing disjoint token spaces as a fundamental barrier to cross-lingual knowledge generalization. Guided by this understanding, we suggest mapping languages into a shared token space by simple word-wise translation and find it substantially improves cross-lingual knowledge generalization, recovering up to 12.6% of native-language learning efficiency --- 14$\times$ the baseline.

Large language models (LLMs) suffer from a long-tail deficit: culturally specific facts, particularly those concerning underrepresented regions such as Latin America, appear too rarely in pretraining corpora to be reliably memorized. Retrieval-Augmented Generation (RAG) addresses this by grounding generation in external text, but structured alternatives such as Knowledge Graphs (KGs) offer tighter control over what enters the context, along with potential gains in explainability and updatability. We benchmark Graph-RAG against standard RAG on LatamQA, a culturally grounded multiple-choice dataset spanning eight thematic categories. The graphs are built end-to-end from Wikipedia articles with KGGen, a recent open-domain extractor, without manual curation in our main setting. G-Retriever is competitive with RAG and reduces the error of the base LLM by 72% with a standard KG and 78% with a benchmark-aware variant, the gap to RAG narrowing further as the graph is oriented toward task-relevant content. The trained projection transfers zero-shot to Portuguese without target-language fine-tuning, indicating multilingual reach.

Table understanding is a core task in document intelligence, encompassing two key subtasks: table reconstruction and table visual question answering (TabVQA). While recent approaches predominantly rely on vision- language models (VLMs) operating on table images, we propose a more scalable and effective alternative based on structured textual representations. These representations are easier to process, align more naturally with LLMs, and eliminate the need for language-specific visual encoders, making them particularly suitable for multilingual documents. We present DELTA, which separates physical structure recognition, logical structure recognition, and OCR to extract both layout and content accurately. DELTA outputs tables in Optimised Table Structure Language (OTSL), a compact and unified format that encodes cell arrangements and textual content. On table structure recognition (TSR), DELTA achieves TEDS- Structure scores comparable with state-of-the-art methods across FinTabNet, PubTabNet, and PubTables-1M. We further establish its robustness on non-English tables through our curated Hindi benchmark, TORQUE. Building on this, we introduce TARQA, an LLM fine-tuned on OTSL sequences. Our approach yields gains of 9.3 p.p. on WTQ (TabQA) and 9.2 p.p. on FinTabNetQA (TabVQA), respectively. On TORQUE, our method ranks second among all VLMs and DELTA + LLM variants. We release our code, models, and benchmark at: https://github.com/Tihiitborg/Tables-Decoded

Clinical de-identification relies on accurately identifying personally identifiable information (PII). However, manually annotated datasets are costly to construct, while existing synthetic alternatives often provide limited details about their generation process or rely on relatively simple synthesis strategies. We introduce Meddies-PII-Dataset, a corpus of one million synthetic clinical documents spanning seventeen languages and nine PII labels. The documents are generated using attribute-conditioned prompts and validated through thirteen deterministic gates that enforce structural and annotation consistency. To evaluate the dataset's utility, we train Meddies-PII-Model, a BIOES token classifier, and compare it with existing PII extraction systems using exact-match entity-level F1. Meddies-PII-Model achieves the highest performance among the evaluated systems on all reported benchmarks, with a mean F1 of 0.827 across fifteen external benchmarks, compared with 0.658 for the strongest baseline. Upon acceptance, we will publicly release the dataset, benchmark suite, model, generation framework, and evaluation code to support research on multilingual clinical de-identification.

Russian state propaganda spreads across many languages and online spaces. Yet, most computational work examines only one such space, usually social media, in one or two languages, and analyses sources rather than content. We introduce SWARM (Search-Web documents Annotated for Russian propaganda, Multilingual), a dataset of 2,183 search engine results across nine languages and diverse web domains (e.g., news, blogs, government sites), each annotated by trained coders for whether it supports a recurring Russian propaganda narrative. We benchmark a source-based blocklist, supervised classifiers, and zero-shot LLMs against these labels. The blocklist misses most propaganda-supporting documents, because such content is not confined to flagged "propaganda" outlets but also appears on mainstream ones. Content-level analysis helps, though how much depends on the model: the strongest LLM reaches a positive-class F1 of 0.73, whereas the supervised classifiers reach only about 0.5, with the smaller LLMs over-predicting support, mistaking topical relevance for endorsement. Detecting search-borne propaganda thus requires per-language, content-level evaluation, which we hope SWARM and our evaluation code enable.

Multilingual large language models (LLMs) are increasingly used for open-ended text generation, yet their behaviour in low-resource languages remains poorly understood. In this work, we question how correct and reliable is the generation of multilingual LLMs when used for the task of story generation. We consider Urdu language as a representative low-resource language. We generate Urdu-Stories, a corpus of 93 stories generated using three contemporary LLMs (GPT-5.1, Qwen-3-Max, DeepSeek-3.1). We manually annotate the errors present in them under a nine-label linguistic, semantic, and cultural taxonomy. Our notable findings suggest that LLMs often make basic errors of grammar and semantics. The stories lack coherence, have unnatural repetition and show pervasive cultural shallowness. We further show using few-shot prompting that the cultural and context errors largely remain unresolved. Our findings highlight the limitations of current LLMs as a reliable source of content generation and information retrieval for low-resource languages.

Transformer-based models excel at Automatic Readability Assessment (ARA), yet feature-based models remain in active use because their predictions tie back to linguistic properties. This matters because readability labels are subjective and rater-dependent, so high accuracy on noisy ground truth may reflect surface patterns rather than the linguistic structure that defines difficulty. We test whether transformers internalize the same features as traditional models across Arabic, English, French, Hindi, and Russian using the ReadMe++ dataset. Shapley Additive Explanations (SHAP) identify the features driving traditional classifiers, which we then use as TCAV concept sets to probe multilingual XLM-R and language-specific encoders. Transformers recover surface-length, syntactic, and lexical-diversity signals, and reflect the ordinal CEFR structure of the traditional models. Alignment varies by model family, language, and layer, with language-specific encoders tracking traditional models more clearly than XLM-R. High linear separability does not always imply directional influence, limiting linear probing for count-based readability features.

Reasoning language models have made substantial advances on a variety of complex tasks, yet their capabilities remain overwhelmingly English-centric: models primarily reason in English regardless of the language they are prompted in. This is inaccessible for non-English-speaking users, risks losing the intent of the original question, and forgoes knowledge more readily expressed in the target language. In this work, we advance L2 reasoning, the ability of a model to reason consistently in the language of the user's prompt, thus building an in-language bridge between the prompt and the answer. We approach this problem from a data-centric angle, investigating how to optimize data composition and scheduling in SFT for reasoning generalization. Building Tiny Aya L2-Thinker at 3.35B scale, we achieve an L2 reasoning rate above 93% across 60 languages on 6 benchmarks spanning math, commonsense reasoning, instruction following, open-ended generation, and cultural reasoning while keeping performance strong. We show the path to generalizing L2 reasoning to held-out languages goes through broader language coverage, readily available multilingual non-reasoning data, and a sufficient English reasoning backbone. These findings indicate that reasoning is a language-agnostic behavior that can be transferred across typologically diverse languages through careful data mixing and without requiring reasoning supervision in every target language. We release our model weights and multilingual reasoning data to support further research on accessible, in-language reasoning.

This study examines the compositionality of steering vectors for language and behavioral control in large language models. Focusing on language, jailbreak, and conciseness, we investigate whether additive, training-free composition of attribute steering vectors can preserve the intended steering effect of each attribute, across four instruction-tuned models from two model families and two size scales. We find that single-attribute steering is reliable for all three attributes, but only within an appropriate combination of intervention layer and steering strength, with abstract behaviors (jailbreak, conciseness) favoring middle layers and language favoring earlier layers. We show that additive composition of two attribute vectors succeeds in steering both attributes simultaneously when each is injected at its own best-performing layer, and that this partially extends to three simultaneously composed attributes, addressing an inconsistency left open by prior work on training-free composition. We further analyze the geometric properties of these steering vectors, finding that they are approximately orthogonal in the residual stream, consistent with their compositional behavior.

Aligned language models fail under two independent pressures: the structural jailbreak class recently formalized as Involuntary In-Context Learning (IICL), which reframes a harmful request as the final missing cell of a data-labeling task completed by pattern rather than judged as content; and the erosion of safety alignment outside English. A natural hypothesis is that these compound. We test it directly. Using a deterministic IICL operator and a StrongREJECT-style rubric judge, we red-team two Google Gemini models on two benchmarks, a 30 general-harm behaviours from HarmBench and 30 financial-abuse behaviours from FinProof, each under a single-shot baseline and under IICL in four languages (English, Spanish, Hindi, Arabic). First, IICL generalizes to a second provider and is worse in finance: it lifts attack success from <=6.7% to 80-90% on HarmBench and 97-100% on FinProof, an order of magnitude above the <=24% its introducing study reported on OpenAI's GPT-5.4. Second, against the hypothesis, forcing the IICL output into a non-English language does not stack the two weaknesses, it attenuates the attack. Eleven of twelve non-English conditions score below their English baseline (sign test, p~0.003), the lone exception a ceiling tie near 100%; on the stronger model's financial set Arabic collapses from 100% to 33%. We attribute this to a relevance curse: once structure has unlocked compliance, the models produce lower-quality harmful content in lower-resource languages, which a substance-grading judge scores as partial. The pattern replicates under an independent non-Google judge (Cohen's kappa=0.86, 377 paired verdicts), and 76.6% of non-English responses were verified in-language. Jailbreak vulnerabilities are therefore not additive; the dominant residual risk is the English structural attack, most acute for financial abuse, not a multilingual one.

Building language technologies and conducting NLP research for low-resource languages---particularly when led by native speakers or involving participatory research practices---are often framed as means of addressing inequality, serving local communities, and, at times, contributing to decolonisation. In this paper, we examine recently published NLP and ML papers, focusing on the narratives used to characterise multilinguality, low-resource languages, and underrepresented cultures. We propose a framework for analysing research framings and identify recurring rhetorical patterns that may hinder accountability and constrain equitable knowledge production for---and by---underserved communities. We further assess the evidential basis of assertions regarding community benefit and find that such statements are often weakly supported or left unsubstantiated. Although community ownership and participation are frequently presented as key objectives, our analysis, supported by statistics from the ACL Anthology, suggests that research outputs more often prioritise resource creation and benchmarking---important but distinct goals---over evidence of broader structural change. We conclude by offering practical recommendations to help authors, reviewers, and readers critically assess these assertions and avoid potentially misleading framings.

Recent multilingual vision--language encoders cover hundreds of languages in a single model, yet on two state-of-the-art instances retrieval on low-resource languages (LRL; e.g. Swahili) trails high-resource ones (HRL; e.g. English) by $30^+$,pp. We ask where in the trained encoder this gap is located. Prior modality-gap and cross-lingual subspace work suggests a linear language direction at the output crowds out alignment-relevant geometry. We falsify this: LEACE drives the linear language classifier from $>99\%$ to near chance and iterated INLP to $37$--$50\%$ while LRL retrieval moves within $\pm 1.5$,pp and all tier means within $2.2$,pp, tracking random controls. The linear bias is a \emph{symptom}, not the cause. Instead, the alignment-causal factor lies along the encoder's forward path: the EOS (end-of-sequence) hidden state's per-language trajectory diverges with depth. Substituting the EOS with its parallel English value three blocks before the projector lifts Swahili from $22.1\%$ to $69.1\%$ on one encoder (and reproduces on the other); three controls rule out pooled-position tautology and English specificity. A front-layer trunk that pulls each language's projection toward the parallel-content centroid corroborates the diagnosis at training time, recovering $+9.6$ / $+17.1$,pp on LRL XM3600 retrieval (1{,}000-image subset), with consistent gains across three further benchmarks while preserving HRL performance.

Solvability detection is one of the most challenging aspects of mathematical reasoning for Large Language Models (LLMs). While prior work has studied this capability extensively, these analyses have been limited to English. Consequently, it remains unclear whether multilingual failures arise from differences in internal Solvability Belief or from language-dependent failures to express it. To address this gap, we introduce the first multilingual benchmark of paired solvable and unsolvable mathematical problems, extending ReliableMath to French and Greek. Using this, we train multilingual probes predicting Solvability Belief and analyze the solvability detection capabilities of state-of-the-art LLMs behaviorally, representationally, and in terms of faithfulness. We find that Solvability Belief is encoded as a largely universal, language-agnostic feature, and that higher-resource languages such as English, despite achieving stronger mathematical reasoning performance, exhibit lower solvability-detection faithfulness.

We present Arkios, a 1.04B-parameter dense transformer pretrained from scratch on 150B tokens of bilingual English-Nepali text, using a custom single-file C/CUDA training stack and a Devanagari-aware byte-level BPE tokenizer built for this project. On ARC-Easy and ARC-Challenge, Arkios exceeds three comparably sized open models (Pythia-1.4B, TinyLlama-1.1B, OLMo-1B) despite an order of magnitude fewer training tokens, likely aided by a match between our educational-web-text pretraining data and ARC's grade-school-science format rather than a general capability advantage. We report full evaluation results under standard protocols, including a correction to an earlier partial-sample estimate, and findings specific to evaluating small models in a low-resource language: the standard multiple-choice-letter prompt format used by common evaluation harnesses places this model at chance on Nepali reading comprehension, and simultaneously at chance on English in the same format, which would lead a naive benchmark run to conclude the model has no Nepali ability when in fact it does. Concretely, both languages score at chance in the letter-choice format (0.240 Nepali, 0.236 English, against a chance baseline of 0.250), while scoring the answer text directly reveals genuine, English-favoring comprehension (0.306 Nepali, 0.387 English). We describe a manifest-conditioned tool-use contract introduced during instruction tuning, where tool calls are permitted only when a tool manifest is declared in context and suppressed otherwise, and report where that contract holds and where it does not. We release both the base and instruction-tuned model weights under Apache-2.0. The training code and a small privately-sourced portion of the Nepali pretraining corpus are not released; everything needed to reproduce the reported numbers from the released weights is included here.

A concept can carry different associations across languages, while modern language models learn English alongside many other languages during pretraining. Yet comparisons among existing models cannot easily isolate how any one language changes the way these models represent English concepts because their training corpora, compute, architectures, and random seeds all differ. We study this question through a controlled experiment with 40 matched 310M-parameter decoder-only models that share an architecture, tokenizer, training recipe, and English data source. Each bilingual condition adds one of eight languages, while four experimental comparisons separately account for English exposure, total training, and English-document overlap. We align each model pair using 3,000 common English words, then measure where 1,000 held-out English concepts fall along 50 fixed semantic contrasts, such as red versus white. Across 32 experimental comparisons, English concept positions differ more between bilingual and English-only conditions than between English-only runs with different random seeds. These differences are larger in contextual states than in token embeddings and peak in middle layers. The language learned alongside English can therefore change how a model represents English concepts even when its English input representations are explicitly aligned.

A central goal of multilingual NLP is to achieve high monolingual performance per language and cross-lingual alignment for large-scale language coverage with a multilingual model. The curse of multilinguality describes the phenomenon of degradation in multilingual model performance as we increase language coverage, posing a threat to the above goal. This paper asks whether multilingual embedding spaces are inherently incapable of achieving perfect multilinguality without a prohibitive increase in required capacity. We first formalize the goal of "perfect multilinguality", embodied in two multilinguality conditions. We then prove that the minimum dimensionality required for perfect multilinguality grows only logarithmically in the number of languages. That is, we show that there is no theoretical curse of multilinguality for embedding space structure. This suggests that the empirical curse of multilinguality is a result of real world data and training conditions. We back this understanding with a small-scale empirical study. Our paper provides the first theoretical and intrinsic perspective on the curse of multilinguality.

The reliability of Large Language Models (LLMs) for API calling degrades in multilingual settings. A common failure occurs when a model selects the correct tool but generates argument values in an inconsistent language, which we term Argument Language Mismatch (ALM). Although semantically correct, such outputs are operationally invalid and not captured by standard API-calling metrics. We revisit post-training strategies for mitigating ALM and find that, in our benchmark, supervised fine-tuning (SFT) provides a strong baseline, substantially improving argument language consistency and end-to-end function call accuracy. Under consistent model selection, SFT achieves performance comparable to, and sometimes exceeding more complex reinforcement learning (RL) approaches. We further examine whether RL with structured, argument-aware rewards offers additional benefits. While methods such as Group Relative Policy Optimization (GRPO) can improve language consistency and better preserve general reasoning ability, these gains are incremental and most pronounced in generalization and multi-objective trade-offs. Overall, our results suggest that much of the performance in multilingual API grounding can be achieved through careful supervised training, with RL providing targeted rather than fundamental improvements.

Agentic benchmarks aim to measure how well AI agents plan, search, execute, and recover within realistic multi-tool environments, but they are almost exclusively in English. As AI agents are globally deployed to a linguistically diverse user base, whether agentic competence measured in English transfers to other languages remains an open question. We introduce OmnilingualGAIA2, a machine-translated expansion (with partial human- expert validation) of the GAIA2 agentic benchmark, covering ten target languages spanning five writing systems, paired with a localised and human-calibrated multilingual verifier. Evaluating seven frontier and open-weight agents, we find a universal cross-lingual gap of 8.8-18.4 pass@3 points that is agent-asymmetric in magnitude, concentrates on tool-orchestration rather than quantitative reasoning, and does not close with model scale. A stratified error attribution decomposes the gap as predominantly model-driven (55%), with a bounded translation-contamination floor of only 6.4% of scenario-language pairs. Human-expert linguistic analysis further identifies morphological cue loss and amplified ambiguity as the primary failure mechanisms in non-Latin-script languages. Our results argue that multilingual agentic evaluation must become a standard part of the reporting protocol for globally deployed agents.

Multilingual neural machine translation models such as NLLB-200 cover 200 languages but leave thousands unsupported, including most Grassfields Bantu languages of Cameroon. When fine-tuning these models for an unseen language, practitioners must choose a proxy language token, yet no principled method exists for this selection. We implemented an embedding initialization strategy where a language token is the average of embeddings from multiple typologically related languages already in the mod el. We evaluate this approach on Limbum-to-English translation using a parallel corpus of 8,837 sentence pairs from New Testament text and a bilingual dictionary. We compare models: NLLB-200 zero-shot (chrF2++ = 12.5), a Transformer trained from scratch (chrF2++ = 14.5), NLLB-200 fine-tuned with a Swahili proxy token (chrF2++ = 47.3), and NLLB-200 with our averaged embedding initialization (chrF2++ = 46.7). We find that the multi-language initialization achieves performance comparable to the best single-language proxy. Both NLLB-200 variants improve over the from-scratch baseline by over 32 chrF2++ points. These results show that multilingual transfer is the dominant factor in extremely low-resource Bantu translation while eliminating the need for heuristic proxy selection. However, all systems fail to preserve tonal diacritics, highlighting an open challenge. We make our dataset and code available to support further research.

Organizing thousands of unstandardized, multilingual expertise declarations is a persistent challenge for Human Resources (HR) platforms, directly impacting downstream tasks like accurate talent matching. To address this, we propose a hybrid knowledge graph generation pipeline that grounds a Large Language Model (LLM) in the Wikidata multilingual Knowledge Graph (KG) while employing an agentic reflexion pattern to synthesize emerging concepts and their associated metadata. Unlike rigid top-down methods or fragmented bottom-up approaches, our system anchors recognized concepts to stable Knowledge Graph entities while dynamically creating new nodes and relational metadata for unrecognized skills. Executed across five stages, entity reconciliation, multilingual canonicalization, active curation, deduplication, and the iterative recovery of unmapped concepts, the system autonomously adapts to rapidly evolving, noisy skill mentions across five European languages. Ultimately, this pipeline provides a highly scalable, explicable, and self-healing framework for generating a comprehensive skills knowledge graph, from which a structured taxonomy is derived, using unstructured, noisy text.

LVLMs are increasingly used to read financial charts, tables, and documents, where a single misread figure can move a decision and the most authoritative-looking answer is sometimes one the model produced without reading the exhibit. The operational question is therefore trust, not accuracy: which answers can be acted on, and which escalated to a reviewer. We evaluate seven confidence estimators, three inference-only and four trained internal probes, across five open-weight LVLMs and four conditions from three financial visual question-answering benchmarks, one bilingual; every probe is trained only on natural images and applied to finance without adaptation, so the results measure out-of-distribution transfer. Three findings hold. First, the scarce property is calibration, not ranking: the inference baselines rank correct above incorrect answers competitively but are badly overconfident, calibration error far above what a threshold can tolerate, and only the trained probes produce a thresholdable score. Second, reliability is structured rather than global, along two axes a practitioner can read directly: the best estimator shifts with both model and task, none leading more than eight of twenty (model, condition) cells, and a controlled bilingual contrast exposes an apparent language robustness as a composition artifact that dissolves once models are read one at a time. Third, cast as deferral under an error budget, how much can be safely automated is set first by the model's competence and only narrowed by its confidence, so deferral clears a real share of the easiest condition and almost none of the hardest, near zero at a strict 5% budget. Two trained probes carry the calibration a deferral policy needs, and among them only the grounding-aware one lowers its confidence on answers a model gives without using the figure, separating detected non-grounding from a fluent guess.

Multilingual text embedding models are commonly adapted using a single training objective across diverse tasks, despite different tasks requiring fundamentally different optimization strategies. We introduce Task-Conditional Flow Matching (TCFM), a multilingual embedding adaptation framework that selectively applies Flow Matching to translation tasks while optimizing retrieval, classification, and pair-classification tasks with objectives better aligned to their learning dynamics. TCFM further combines teacher-guided representation preservation with a three-stage curriculum to enable stable adaptation. Evaluated on the Indic Massive Text Embedding Benchmark, TCFM establishes a new state-of-the-art, consistently improving embedding quality across a diverse set of multilingual tasks and generalizing across embedding model families. We will publicly release the codebase and datasets upon acceptance of the paper.

Multimodal large language models (MLLMs) have achieved significant success in speech-to-text translation (S2TT). However, when processing multilingual speech inputs, a single speech encoder shared across all languages suffers from the curse of multilinguality: languages at different resource levels compete for limited representation capacity, leading to strong high-resource performance but substantial degradation on low-resource speech. To address this problem and improve multilingual consistency, we propose MSRT, a novel framework built around a resource-aware Mixture of Speech Encoders (MoSE). MoSE uses an explicit language router to assign each utterance to an appropriate expert encoder. A frozen expert preserves high-resource language capabilities, while a trainable expert adapts to and specializes in medium- and low-resource languages. We further introduce a five-stage curriculum learning strategy that substantially reduces data dependence, requiring only 10 hours of paired S2TT data per language for effective alignment. We conduct extensive experiments on 45 languages, systematically evaluating all $45 \times 44$ translation directions. Our 4B-parameter model achieves state-of-the-art performance, outperforming substantially larger baselines. Empirical analyses show that MoSE improves high-, medium-, and low-resource languages simultaneously, with the largest gains on low-resource speech, thereby breaking the curse of multilinguality without compromising high-resource performance. To support future multilingual S2TT research, we release our code and models.

Multilingual language models are deployed across a hundred or more languages, yet most benchmarks test whether a model can perform a task in a language rather than whether it commands the language itself, conflating fluency with proficiency. We introduce M-GATE (Multilingual Grammar, Accuracy in Translation, and Efficiency), a benchmark of linguistic proficiency spanning 30 typologically diverse languages from high- to low-resource. M-GATE comprises three tasks: grammatical error detection on linguist-crafted, adversarially selected sentences that turn on hard, language-specific phenomena; round-trip translation of shared English sources across 29 target languages, scored by a three-provider LLM judge panel validated against professional annotators; and a supplementary tokenizer-efficiency measure. We evaluate over 50 models in more than 80 configurations. Fluency and proficiency come apart sharply: models that translate competently sit near chance on the adversarial grammar items, the best reaching a Matthews correlation coefficient (MCC) of only 0.36, and their errors lean systematically toward under-flagging, accepting ungrammatical text rather than raising false alarms. Translation quality closely tracks a language's share of pretraining data (r = 0.86 against log Common Crawl share), producing a steep low-resource penalty that is nonetheless narrowing with successive model releases. Enabling reasoning reliably improves translation, while its effect on error detection is smaller and for some models negative, so the best configuration is task-dependent. To resist contamination, test items are kept private behind a continuously updated public leaderboard, with illustrative examples released (https://m-gate.ai).

Multilingual multi-agent systems exhibit substantial degradation beyond English, yet prior work rarely identifies how task-critical information is lost when user requests are converted into executable plans. We study the planner in a multi-agent system as the request-to-action interface and derive an actionable taxonomy of planning-grounding failures from failed real-world task executions. LLM-based analysis shows that these failures constitute an increasing share of unsuccessful executions as language-resource availability declines, with the strongest effects in low-resource languages. To test whether the taxonomy supports mitigation, we introduce TART, Taxonomy-Guided Actionable Representation, that makes the taxonomy's key aspects explicit to the planner and downstream sub-agents. Across multiple languages, three LLM backbones, two datasets, and two agentic configurations, TART consistently improves performance. On multilingual GAIA, it raises a state-of-the-art system's accuracy by 5.6 percentage points averaged across eleven languages spanning low- to high-resource settings.

Modern LLM-based ASR systems have established multilingual capability as a standard feature, leveraging large-scale multilingual corpora and LLMs' cross-lingual knowledge to achieve competitive performance across multilingual benchmarks. However, jointly modeling languages with heterogeneous acoustic, phonological, and lexical characteristics inevitably introduces optimization conflicts, undermining language-wise specialization. To address this challenge, we propose Language-Specialized Multi-Teacher On-Policy Distillation (LS-MOPD), which decouples language-specific knowledge acquisition from multilingual capability integration: language-specialized teachers are independently optimized via reinforcement learning (RL), with their expertise then integrated into a generalist multilingual student through language routing and token-level multi-teacher distillation, thereby reducing direct cross-lingual optimization conflicts. We further explore static and dynamic acoustic-prefix configurations to examine how teacher-student prefix consistency influences the efficacy of on-policy distillation. Experiments on benchmarks covering Mandarin, Mandarin subdialects, Cantonese, and English demonstrate that LS-MOPD substantially outperforms RL baselines and surpasses the empirical performance envelope defined by the best-performing RL teachers on nearly all benchmarks, revealing its potential to generalize beyond all teachers in multilingual ASR.

Full-duplex spoken dialogue systems (FDSDSs) can listen while speaking, enabling natural behaviors such as smooth turn-taking, backchannel handling, and user barge-in handling. However, fair comparisons in multi-turn conversations remain a challenge. In addition, existing benchmarks provide limited coverage of languages and dialogue domains. We propose M3-DuplexBench, a multi-turn, multilingual, multidomain benchmark for FDSDSs. M3-DuplexBench supports English and Japanese and covers both casual conversation and multi-turn question answering. In addition, we evaluate models under multiple dialogue context settings, including single-turn, user-only, and teacher-forced full-context settings, to analyze how dialogue history affects model behavior. Experiments with recent FDSDSs reveal model-specific turn-taking characteristics, clear performance gaps across languages and domains, and mixed effects of dialogue context.

As large vision-language models (LVLMs) are deployed globally, the combination of multilingual instructions and visual information makes malicious attacks more covert and sophisticated than ever before. However, existing methods isolate language and modality defenses, which, coupled with the scarcity of safety data and high fine-tuning costs, makes it difficult for models to defend against compound attacks. To address this severe challenge, we propose a neuron-level cross-dimensional safety alignment framework driven by modality- and language-shared safety neurons (MLS-Neurons). First, we identify monolingual and unimodal safety neurons by comparing responses to harmful and benign samples, quantifying functional saliency through activation strength and downstream impact. Then, by intersecting these unimodal neurons within each language, we extract modality-shared safety neurons (MS-Neurons) responsive to both visual and textual risks, bridging the safety representation gap between modalities. Furthermore, using English as a semantic anchor, we intersect MS-Neurons across languages to identify modality- and language-shared safety neurons (MLS-Neurons), serving as key defenses against compound attacks. Finally, we update only this minimal subset of shared neurons (~0.03% of parameters), transferring English-only safety supervision to multilingual and multimodal scenarios. Extensive experiments show that our method significantly outperforms state-of-the-art approaches across diverse multilingual and multimodal safety benchmarks while preserving general utility.

State-of-the-art retrieval models increasingly rely on closed training data, creating a reproducibility gap. We present an open end-to-end recipe for training retrieval models and study how English supervision transfers to multilingual retrieval through translate-train. We first reconstruct and curate 665M English contrastive pre-training pairs from 1.4B pairs across 34 public sources and build 1.88M supervised fine-tuning pairs with mined hard negatives. Training yields two 149M-parameter models: DenseOn, a single-vector dense model, and LateOn, a ColBERT-style late-interaction model. They achieve 56.20 and 57.22 average nDCG@10 on BEIR, respectively, setting new state-of-the-art results for this size class. We then translate the validated English data into eight languages, yielding 2.8B pairs with cross-lingual samples, and train mDenseOn and mLateOn, two 307M-parameter models built on mmBERT-base. Despite sharing their backbone, data, and objectives, their representations behave differently: the dense model is strong on English and translated languages but degrades outside translate-train support, whereas the late-interaction model generalizes better to unseen languages and scripts. This suggests that token-level matching turns translate-train from a target-language expansion strategy into a multilingual generalization recipe. We publicly release the models, datasets, and training code.

Despite the existence of exponentially many valid tokenizations for a given string, language models operate on a single canonical sequence deterministically produced by the tokenizer, leaving the broader tokenization space largely uncharacterized. In this paper, we investigate this overlooked space by studying the behavior of language models under non-canonical tokenizations across diverse languages. For English, prior work shows that models are largely invariant to alternative tokenizations that represent the same underlying string. We ask whether this invariance generalizes to other languages beyond English. We conduct a multilingual study across 27 languages spanning diverse scripts and evaluate LLM behavior under alternative tokenizations across six downstream tasks. We find that tokenization invariance does not generalize: model behavior varies substantially across languages with instruction-tuned models exhibiting an average relative performance drop of 23.7% for Llama-3.1-8B, 11.4% for Qwen3-8B, and 9.9% for Gemma-3-12B. The variation of tokenization invariance is systematic across languages. Languages that exhibit higher token fragmentation show significantly greater sensitivity to non-canonical tokenizations. Our study of tokenization robustness serves as a diagnostic of how tightly a model is coupled to its tokenizer. These results demonstrate that tokenization robustness is not a universal property of language models, but depends strongly on the language and its interaction with the tokenizer. We also show that LoRA fine-tuning with multi-tokenization training data provides an effective mitigation for tokenization sensitivity. Fine-tuning on English alone improves tokenization robustness across languages, while systematically sampling diverse non-canonical tokenizations achieves the strongest overall performance.

Offensive language detection models generally suffer performance degradation when deployed across datasets and across languages, yet most existing studies stop at reporting this phenomenon and lack a systematic methodology for decomposing the causes of degradation into attributable components and quantifying the cost of remediation. This paper proposes a diagnosis and optimization framework composed of three coordinated technical components. First, a zero-shot transfer loss decomposition that separates the performance degradation from OLID to MLMA into two independently measurable components, namely dataset effect and language effect. Second, a controlled fine-tuning protocol that quantifies both adaptation efficiency and the hidden damage inflicted on the source task by comparing few shot learning curves under continued fine-tuning and cold-start starting points. Third, three joint training strategies incorpo rating temperature sampling and experience replay, which offer a controllable Pareto trade-off between improving multilingual capability and preserving source-task performance. Experiments built on this framework show that the dataset effect dominates the zero-shot transfer loss and substantially outweighs the language effect. Few-shot adaptation without a replay mechanism, though data-efficient, inflicts source task damage 4 to 9 times greater than that of the joint training strategies, and its damage magnitude is highly unstable. The three joint training strategies trade 3.2 to 4.1 percentage points of source-task performance for 8.1 to 42.6 percentage points of multilingual capability gain, forming a clear and controllable Pareto trade-off.

Instruction tuning has become the standard method for adapting large language models to follow human intent, yet existing instruction datasets are dominated by English-language general-knowledge tasks and lack coverage of specialized pedagogical domains. This paper presents IKS-Instruct, a dataset of 24,795 instruction-response pairs for teaching language models to deliver educational content grounded in Indian Knowledge Systems (IKS). The dataset spans seven languages (English, Hindi, Sanskrit, Tamil, Telugu, Kannada, and Malayalam), covers 41 pedagogical techniques from the Vedic oral and mathematical traditions, and is aligned with the Central Board of Secondary Education (CBSE) curriculum for classes 6 through 12. The pairs are derived from six source types: classical text corpora (Bhagavad Gita, Thirukkural, Sangam literature, Vedic texts), curriculum-aligned pedagogical templates, Vedic mathematical sutra demonstrations, bilingual instruction pairs, technique-grounded multi-turn dialogues, and cross-tradition comparative analyses. Quality is assessed through a multi-judge evaluation framework in which independent language models score responses on 12 dimensions including technique fidelity, pedagogical quality, factual accuracy, and IKS cultural depth. Under a uniform five-judge external panel (median aggregation over 1,201 stratified items), the strongest IKS-Instruct fine-tune of a compact 7B model reaches a median judge score of 6.39, within 0.15 of a strong general-purpose reference model (Nemotron-Nano at 6.54) at a fraction of its deployment cost, while the base model without IKS fine-tuning scores near zero on the IKS-specific dimensions. Model quality does not increase monotonically with data curation, a result we report together with the corresponding data-quality gains.

Grievance is one of the warning signs analysts look for when assessing threats of violence. It is increasingly measured at scale from online text, most often with word-level lexicons like the Grievance Dictionary that score by matching weighted terms. Such matching is a fast and transparent proxy, but it cannot resolve whether a term is asserted, quoted, negated, or condemned. These lexicons are also often evaluated on pools enriched with the very examples they retrieve, so a high score partly reflects agreement with the lexicon's own selection rule. Examining a five-language, 2{,}000-item evaluation pool, we find its halves separated almost perfectly by the lexicon itself: every item labeled ``random'' is in fact lexicon-negative, so the lexicon's apparent macro-AUROC of 0.686 collapses to a 0.500 floor fixed by construction. We keep the dictionary's 22-construct ontology but replace term matching with context-reading models, evaluated on a non-circular benchmark that separates unconditional-random, lexicon-positive, and lexicon-negative strata across five languages. Reading the full post rather than the target sentence alone helps most where the lexicon is silent, raising average precision on lexicon-negative text from 0.14 to 0.20, with the largest gains on quoted, implicit, and cross-sentence grievance. Together, these results show that grievance is measured more faithfully by reading the surrounding context, and more honestly when tested on text the lexicon did not select. We release our code and benchmark at https://github.com/behavioral-ds/multilingual_grievance.

Although Large Language Models (LLMs) demonstrate remarkable multilingual fluency, their internal knowledge representations remain disproportionately biased toward high-resource languages. This leads to cross-lingual factual inconsistency, where they shift their empirical answer distributions based solely on the prompt language. We investigate whether these biases can be mitigated at inference time, forcing an English-prompted model to answer as if it were queried in target languages (German, Spanish, Bulgarian), and evaluate four intervention strategies: zero-shot contextual steering (persona prompting), internal representation manipulation via Contrastive Activation Addition (CAA), and lightweight weight modification via Direct Preference Optimization (DPO) trained on benchmark-derived factual data as well as conceptual generalization data. To assess alignment, we curate a multilingual factual dataset alongside a novel generalization benchmark comprising culturally rooted queries to determine whether factual interventions transfer to broader target-centric preferences. Experiments on Gemma 3 12B Instruct reveal persona prompting to be the strongest overall intervention, balancing efficacy, safety, and out-of-domain generalization. While CAA yields sharp inconsistency benchmark shifts, it is configuration-sensitive and risks knowledge degradation. DPO-based adapters offer permanent, yet narrower and less transferable gains. These findings suggest that cross-lingual inconsistency is at least partly a selection problem, and that simple contextual interventions may outperform more invasive methods for robust, transferable alignment.

This paper reports on a collaboration between the Directorate-General for Translation (DGT) and the European Master's in Translation (EMT) to localise the MMLU dataset into 11 European languages. Beyond creating a more inclusive benchmark for LLM evaluation, the project offers master's students authentic, project-based professional training in translation, revision, project management, and multilingual coordination, while highlighting key methodological, administrative, and workflow challenges.

Language models encode text as subword tokens, raw bytes, or rendered pixels, but these encodings are usually compared under modeling constraints that expose different amounts of linguistic content to models across different languages. We instead ask what each encoding preserves when both the content and the downstream capacity are controlled. Using verified parallel sentences across thirteen languages and five scripts, we compare tokens, bytes, and pixels through a shared bottleneck whose width is swept to trace rate-utility frontiers. This separates three quantities that are often conflated: the number of input positions an encoding creates, the latent capacity available after encoding, and the task-relevant information that survives compression. We evaluate three utilities: surface form preservation, cross-lingual sentence alignment, and topic classification. No encoding dominates across tasks or capacity regimes. Pixels preserve surface form best, bytes preserve cross-lingual alignment best, especially in same-script multilingual settings, and tokens support topic prediction best. These performances are not explained by sequence length alone. Short inputs can discard useful meaning, while long inputs can preserve information that compresses well. Choosing an encoding is therefore not a fixed preference for tokens, bytes, or pixels, but a rate-utility tradeoff that depends on the task, language mix, capacity regime, and compute budget.

Moderation systems increasingly rely on external toxicity tools, but those tools are unreliable under code-mixing, transliteration, slang, and language mismatch. We study the \emph{conditional reliability} of toxicity priors in Indian multilingual and code-mixed short text: English toxicity, Indic abuse, and rule-based severity cues can be useful evidence, but only in some linguistic and abuse-severity contexts. We propose ToxGate, a trust-fusion head that conditions each auxiliary signal on the encoder representation before adding it to the prediction state. Across three short-text abuse datasets, four transformer encoders, and five seeds per setting, ToxGate improves over matched plain encoders in 10 of 12 in-domain settings and 7 of 8 transfer settings. The largest and most interpretable gains occur in high-risk moderation slices, including explicit slurs, violent threats, and cross-dataset transfer. The broader lesson is that moderation systems should treat external toxicity tools and priors as conditional evidence rather than fixed features or ground truth, in focused ablations, source-specific gating gives the strongest results in transfer, severe-abuse slices, and high-risk triage.

Critical thinking is a fundamental skill that helps learners move beyond simple memorization. One way to develop this skill is through high-order questioning. However, crafting such questions remains a challenge for educators, and classroom practices tend to rely on low-order questions. Large Language Models have demonstrated strong capabilities in generating high-order questions, especially when guided by prompts based on Bloom's Taxonomy. Yet, existing research has largely centered on this framework and focused only on English. This study addresses these gaps by introducing prompts grounded in two alternative frameworks: Claim-Evidence-Reasoning and Divergent Questioning within a multilingual context using Basque, Spanish, and English. Results indicate that while both an open-source and a proprietary model rather effectively generate questions in all three languages, only about half of the answerable questions are recognized by teachers as high-order. A positive finding is that the alternative frameworks produce structurally and conceptually varied questions, suggesting they could complement each other and provide viable alternatives to Bloom's Taxonomy.

Large-scale, richly annotated career trajectory data underpins workforce planning, job recommendation, and labour market analysis, yet publicly available datasets are either small, closed to independent use, or built from pre-standardized occupational codes with LLM-synthesized rather than authentic free text. We present JobHopv2, an improved version of the publicly available JobHop dataset, constructed through end-to-end large language model (LLM) extraction from a corpus of ${\sim}440{,}000$ pseudonymized, multilingual resumes provided by VDAB, the Flemish Public Employment Service. The released dataset comprises $355{,}315$ career trajectories annotated with ESCO occupational codes, quarter-level temporal information, and normalized five-level education attainment, broadening both the coverage and the annotation richness of the original release. Relative to v1, JobHopv2 introduces a redesigned extraction pipeline based on reasoning-controlled LLM inference with a retry mechanism (achieving a 100% JSON parse rate), a richer extraction schema, and a revised evaluation protocol scored against three complementary annotation baselines. Evaluated against these baselines, our best extractor comes closest to the inter-annotator agreement ceiling among all compared models, trailing it by only 1.1-2.7 percentage points. The dataset and code are publicly released to support reproducible career-trajectory research.

Despite recent scaling successes, multilingual ASR performance remains highly uneven, with long-tail languages suffering from severe data scarcity. This work addresses the challenge of building robust foundation models for underrepresented Central Asian languages (Kazakh, Kyrgyz, Uzbek). We present GigaAM Multilingual, a Conformer encoder pre-trained on 2M hours of audio using a HuBERT-style objective. Crucially, we introduce a cluster-level data balancing strategy during pre-training and a domain-aware sampling method during fine-tuning to mitigate head-language dominance. In controlled comparisons, our approach outperforms strong open pretrained encoders (Whisper Large v3, Omnilingual-1B) on target languages, achieving significant gains on spontaneous speech while maintaining efficiency. We release the foundation encoder and ASR model, offering a proven recipe for effective multilingual adaptation under realistic data imbalance.

The rapid development of large language models and multimodal large language models has accelerated the emergence of proactive agents capable of operating everyday tools and assisting users in real-world environments. However, existing benchmarks struggle to evaluate such agents effectively, as they often rely on sandboxed environments and single-turn evaluation paradigms. Moreover, their scenario-based task taxonomies mix multiple model capabilities within the same task category, making it difficult to identify the root causes of agent failures. To address these limitations, we introduce UniClawBench, the first capability-driven benchmark designed to evaluate proactive agents in dynamic, real-world settings. UniClawBench is built around five foundational model capabilities: Skill Usage, Exploration, Long-Context Reasoning, Multimodal Understanding, and Cross-Platform Coordination. Based on these capabilities, we design 400 bilingual real-world tasks. Unlike previous benchmarks that rely on static, pre-recorded answers, our benchmark evaluates agents in live Docker containers using fine-grained, step-by-step completion checkpoints. Furthermore, we design a closed-loop evaluation strategy comprising an executor agent, a hidden supervisor agent, and a user agent to simulate realistic multi-turn human feedback without leaking grading criteria. To disentangle base model capabilities from framework-level design choices, we evaluate state-of-the-art models under multiple agent frameworks. Through comprehensive comparisons across both models and frameworks, we show how base model capabilities and agent framework designs jointly shape performance in real-world environments. To facilitate future research, we make our benchmark and code publicly available at https://github.com/HKU-MMLab/UniClawBench.

One of the expected abilities of vision-language models (VLMs) is spatial reasoning ability based on a given text and image. To evaluate the spatial reasoning abilities of VLMs, we focus on the use of spatial deictic expressions, which are defined as spatial expressions whose referent is determined by their situational context, such as this'' and that''. To handle spatial deictic expressions, VLMs must jointly reason over language and visual space, grounding context-dependent references in the image's spatial structure. In addition, selecting appropriate spatial deictic expressions across languages requires VLMs to understand the language-specific spatial distinctions encoded by these expressions. In this paper, we develop a benchmark to evaluate the multilingual ability of VLMs to use spatial deictic expressions in four languages. Our experiments using this benchmark reveal that the tested models use demonstratives in a manner different from that of humans, particularly in selecting the appropriate demonstratives based on the distance to the object.

We systematically investigate neural speech enhancement systems, ranging from very small ($\sim$10,k parameters) to medium-large ($\sim$2-5,M parameters), which specialize to acoustic conditions using contextual information such as speaker identity, noise type, speaker gender, spoken language, and SNR. By fine-tuning generalist models on specific data subsets, we find that specializing to a speaker's identity consistently yields the largest gains in estimated speech intelligibility and quality. In contrast, specializing to SNR, noise type, or gender offers only marginal benefits. Crucially, we show that a small model specialized to both a specific speaker and a specific noise type can match or exceed the performance of a generalist model ten times its size. Further, cross-lingual tests reveal that models specialized to a target language outperform multilingual generalists, suggesting that language is a salient feature for specialization. These findings highlight the potential of small, adaptive models for resource-constrained applications like hearing aids, which specialize on-the-fly to contextual information.

Portuguese remains underrepresented in text embedding evaluation, despite being one of the most widely spoken languages in the world. As a result, embedding models are often selected based on English or multilingual metrics, while their effectiveness in Portuguese remains unclear. We present MTEB-PT, a Portuguese benchmark constructed from a subset of MMTEB, comprising 14 existing datasets across Semantic Textual Similarity (STS), classification, retrieval, and reranking. We use this benchmark to evaluate 17 open- and closed-source embedding models under a unified protocol. Our results show that Portuguese performance is strongly task-dependent: multilingual rankings do not reliably predict Portuguese-specific performance across task families, no single model dominates all settings, and models with stronger long-context capacity are particularly advantageous on longer-input tasks such as retrieval and reranking. The benchmark also shows that language-specific fine-tuning still improves model performance in Portuguese, especially on task types that match the adaptation data most closely. To examine this effect, we fine-tune three representative backbone models with Portuguese contrastive supervision and Matryoshka Representation Learning (MRL). These benchmark-informed baselines yield their strongest gains on STS, consistent with the predominantly symmetric supervision used during training, while also improving retrieval and remaining competitive under dimensional truncation. We release the MTEB-PT benchmark, the fine-tuned models, and the training and evaluation code.

Multilingual documents encapsulate rich regional cultures, scientific discoveries, and historical records. Parsing this content into structured, machine-readable formats is critical for unlocking global knowledge. However, existing benchmarks predominantly focus on high-resource languages like English and Chinese, creating an evaluation blind spot concerning model performance on other languages. While recent Vision-Language Models (VLMs) claim support for hundreds of languages, the lack of ground truth makes it impossible to empirically verify these capabilities. To bridge this gap, we introduce MORE, a large-scale benchmark designed for multilingual document parsing evaluation. MORE distinguishes itself through three key dimensions: (1) Unprecedented Scale: It covers 149 languages, making it the most linguistically diverse benchmark to date; (2) Structural Complexity: Unlike previous works, it extends evaluation beyond plain text to include structural elements such as code blocks, tables, and catalogs; and (3) Data Authenticity: All samples are curated from real-world documents via a model-assisted, human-refined annotation pipeline. We evaluate state-of-the-art models using MORE, establishing new performance baselines for long-tail languages and validating the benchmark's effectiveness in diagnosing model capabilities in realistic, diverse scenarios. The MORE dataset will be available at https://github.com/zimoqingfeng/MORE.

LLM-as-a-Judge has become the dominant evaluation paradigm for many natural language generation tasks (albeit mostly in English) due to shortcomings of conventional metrics and high correlations with human judgment. There are now attempts to extend LLM-as-a-Judge to multilingual settings including low-resource languages. However, LLMs have limited proficiency in low-resource languages, and there is often no adequate human validation in these settings. To highlight the scope of the problem and current practices, we explore the use of LLM-as-a-Judge evaluators in ACL Anthology papers focusing on multilingual settings and low-resource languages across a diverse set of tasks. Out of 650 papers mentioning LLM-as-a-judge, only 33 of them focus on low-resource or multilingual settings. Our in-depth analysis of these papers indicates inconsistent evaluation outcomes, a tendency to overtrust LLM judgments in multilingual settings, and the widespread reliance on a single judge model per study. To help the NLP community further, we conclude with a checklist of recommendations about how to use LLM-as-a-Judge in multilingual and low-resource settings

Recent advances in automatic speech recognition (ASR) have explored different sequence models, including Conformer-based models and newer state space models such as Mamba. Although prior work has evaluated these architectures in multiple languages, their effectiveness in African languages remains underexplored. In this work, we evaluate Mamba for ASR on seven South African languages. In monolingual experiments, each model is trained on 50 hours of speech per language, and we compare Mamba to a Conformer baseline of similar parameter scale. Mamba achieves similar recognition accuracy to Conformer while using fewer computational resources and training faster. We further evaluate generalization in this setting and find that both models struggle to generalize to speech that is much longer than what they were trained on. We then study multilingual ASR using Mamba models, where the baseline is pooling all languages together. On top of this, we tested three extensions: training with language-family information by adding both language and language-family embeddings as biases to the downsampled acoustic representations, and multitask learning with a CTC ASR objective and a language identification (LID) head. We find that multilingual training consistently improves performance over monolingual training. However, adding explicit language information does not improve in-domain performance but does improve cross-corpus robustness. We conducted ablation studies in low-resource multilingual settings using 5-hour and 10-hour per-language training data, where we observed gains from using language embeddings and further demonstrated that removing or altering them hurt model performance. Lastly, we analysed these embeddings and find that they do not capture linguistic similarity in a typological sense, but instead act as task-specific control vectors.

Multilingual fluency often invites a stronger assumption: a model that can speak a user's language must also understand the culture encoded by that language. We call this the Illusion of Cultural Alignment. To test this assumption directly, we introduce MSQA, a benchmark of 1,064 natively sourced questions across 11 language groups, five cultural dimensions, and three difficulty tiers. Unlike translated benchmarks, MSQA targets locally grounded knowledge and reduces shortcuts from English-centric cross-lingual transfer. Evaluating 18 LLMs, we find substantial cultural degradation and a pronounced Locality Effect: cultural competence tracks pre-training exposure more closely than general reasoning ability. We further show that common inference-time remedies do not dissolve the illusion. Models remain overconfident on unfamiliar cultural questions, repeated sampling yields unstable rather than reliable correctness, and retrieval augmentation helps unevenly on long-tail facts. These findings indicate that cultural alignment cannot be inferred from multilingual ability alone and requires deeper intervention than calibration, sampling, or retrieval at inference time

SSL speech models capture rich phonetic, prosodic, and acoustic patterns from raw audio, yet how they encode articulatory information across diverse languages remains unclear. Using EMA data from bilingual Finnish-Russian speakers, we evaluate cross-lingual correlations between SSL latent representations and articulatory movements. Models achieve strong prediction performance (Pearson r up to 0.68) even with approximately 5 minutes of training data, with multilingual models outperforming monolingual ones. Intermediate layers encode articulatory features most effectively, and tongue movements are more predictable than lip movements. We also assess the impact of task type (read versus spontaneous speech) and language proficiency, finding higher accuracy for structured tasks and strong generalization across proficiency levels. These results enhance the interpretability of SSL models and show their potential for speech-technology applications.

Romanized Code Mixing (RCM), where bilingual speakers fluidly blend local languages with English in Roman script, has emerged as the dominant form of communication across multilingual communities. While Large Language Models (LLMs) perform strongly on monolingual and native-script benchmarks, their ability to follow instructions and reason over RCM-based content remains largely unexplored. To this end, we introduce the Indi-RomCoM benchmark for facilitating systematic evaluation on Indic Romanized Code-Mixed instructions. Our benchmark spans seven instruction-following tasks, four widely spoken Indic languages, and three controlled code-mixing intensity levels. We extensively evaluate a suite of LLMs covering proprietary, open-weight, and Indic-focused models under zero- and few-shot settings. LLMs consistently underperform on RCM instructions, with performance degrading as code-mixing density increases. Furthermore, reasoning tasks suffer less degradation than detection tasks (e.g., Toxicity) because the generated explanations offer necessary context. We believe Indi-RomCoM helps the community in developing inclusive multilingual systems.

Multimodal Large Language Models (MLLMs) have redefined visual understanding by combining vision encoders with large-scale language models. This unified architecture enables strong performance on tasks like image captioning, visual question answering, and multimodal dialogue, often in zero- and few-shot settings. Their general-purpose capabilities and flexible interfaces make MLLMs a promising foundation for real-world vision-language applications. Assistive AI aims to help users interact with their environments through natural language. These scenarios demand robust visual recognition, contextual reasoning, and multilingual comprehension-capabilities that MLLMs are believed to offer. However, their effectiveness in assistive settings remains to be fully understood. In this work, we explore whether MLLMs can support Assistive AI by evaluating state-of-the-art models on real-world tasks: recognizing everyday objects like currency, answering questions based on scene text, and reading visually presented content across multiple languages. To this end, we developed a system, NetraLink, using a head-mounted GoPro to capture real-world egocentric data, and collected a benchmark covering these assistive scenarios. Our findings provide a comprehensive diagnostic of current MLLMs, highlighting their strengths and limitations in enabling assistive technologies grounded in visual perception and language interaction.

Multimodal misinformation detection is increasingly important because viral posts now combine long multilingual narratives, several images, mixed provenance, and subtle text--image framing errors. Existing benchmarks and methods remain poorly matched to this setting: they usually isolate short captions, single images, binary labels, or one manipulation source, while agentic verification remains costly under realistic evidence search. We present ReMMD, a realistic multilingual multi-image agentic verification framework for multimodal misinformation detection. ReMMD includes ReMMDBench, a real-world multimodal misinformation detection benchmark with 500 samples, 2,756 images, five monolingual languages, two cross-lingual settings, three text-length tiers, multi-image posts, five-way veracity labels, eight distortion labels, evidence provenance, and rationales. It also includes ReMMD-Agent, a persistent-memory verifier that decomposes posts into atomic points, builds a reusable evidence set, and predicts structured L1/L2/L3 outputs. Across proprietary systems, open LVLMs, MMD-Agent, and T2-Agent, ReMMD-Agent obtains the best five-way veracity performance, with 41.80% accuracy and 39.12% macro-F1 using GPT-5.2, while reducing cost by 17.5% relative to MMD-Agent and 79.9% relative to T2-Agent. The project is available at https://dang-ai.github.io/ReMMD.

We present a multilingual computational discourse analysis of how language constructed the algorithmic consecration of Vozinha, the 40-year-old Cape Verde goalkeeper, after Spain 0-0 Cape Verde at the 2026 FIFA World Cup. The study contributes a multilingual corpus in Portuguese, Spanish, English, and French; a nine-frame narrative taxonomy with cue-based frame annotation; a reproducible annotation pipeline combining LLM-assisted suggestion with human validation; and an analysis of cross-lingual narrative diffusion across discourse phases. We treat the platform follower count itself, narrated as "50k to 8M", as a linguistic object: a circulating and narratable proof of visibility rather than a mere measurement. The follower-growth timeline is used only as contextual metadata: we reconstruct a conservative phase structure, not a continuous API-native series, and type every datapoint by value class, confidence, and evidence type. The only exact primary scraper anchor is 8,235,652 followers at 2026-06-16 15:47 UTC; all other figures are reported as estimated ranges or thresholds, including an estimated pre-match baseline of 45k-56k. Findings suggest that distinct languages carried distinct frames: Portuguese mobilization, Spanish crisis, English nation-making, and a shared platform-metric spectacle through which peripheral athletic performance became globally visible. As a v0.1 pilot, the paper releases the corpus schema, frame taxonomy, annotation guidelines, hashed visual-evidence log, and typed timeline, while flagging full double annotation and inter-annotator agreement as planned work.

Current multilingual evaluations for Vision-Language Models (VLMs) assume a one-to-one mapping between language and orthography, overlooking billions of users of multi-script languages. We introduce PuMVR (Punjabi Multimodal Visual Reasoning), a benchmark of 1,000 strictly parallel image-text instances across Punjabi's three active scripts: Gurmukhi, Shahmukhi, and Roman. Evaluating 10 state-of-the-art VLMs, we expose a substantial and systematic Script Gap. Models frequently solve visual tasks in one script while failing identical tasks in another, with accuracy deltas reaching 16%. Crucially, visual input boosts absolute performance uniformly yet does not close the orthographic gap. Furthermore, cross-script in-context transfer is highly brittle, exposing script-locked knowledge representation. Supported by McNemar tests across all script pairs, our findings demonstrate that current "multilingual" VLMs are not truly multi-script. We propose the Script Consistency Rate (SCR), which falls as low as 24.8% on our benchmark, as a mandatory metric for script-agnostic evaluation to ensure equitable AI access. Data and code are available at: https://github.com/prabhjotschugh/Not-Truly-Multilingual-PuMVR.

Symbolic informalization enables a reliable conversion of formal mathematics to natural language. It has the potential to make machine-checked content human-readable without loss of precision. In a traditional proof system usage, symbolic informalization generalizes the limited mechanisms of syntactic sugar into the ordinary language of mathematics. In a setting where proofs are constructed by artificial intelligence and autoformalization, symbolic informalization can explain what precisely has been constructed. This paper outlines the project Informath, which aims to show how symbolic informalization can produce fluent text with a reasonable development effort and address multiple formal and natural languages. Informath is based on an interlingual architecture, where Dedukti works as a hub between different proof systems (Agda, Lean, Rocq) and Grammatical Framework (GF) takes care of linguistic correctness and variation in different natural languages.

SMS fraud is increasingly cross-channel: a message directs the user to a webpage, and the final risk depends on how the SMS claim aligns with the page content and requested user action. However, existing evaluations either focus on message-only smishing classification or expose URL and domain cues that allow models to rely on reputation shortcuts. To address this gap, we introduce \textbf{FraudSMSWalker}, a controlled benchmark for URL-masked SMS-to-webpage fraud judgment. FraudSMSWalker contains 699 bilingual chains, including 332 fraudulent and 367 benign cases, across ten service scenarios. The model-visible input consists of the SMS context and sanitized webpage evidence, while raw URLs, hosts, domains, IPs, redirects, and reputation metadata are withheld. The benchmark further includes hard benign cases whose pages contain login, payment, verification, or account-management elements that are plausible under the service context but also appear in scam flows. We evaluate nine web agents under masked browser-agent protocols and conduct URL-visibility ablations. The results show that current agents can detect suspicious cues, but struggle to preserve benign recall and often produce positive predictions that are weakly supported by the observed evidence. These findings position FraudSMSWalker as a benchmark for measuring whether web agents can make fraud judgments that remain both accurate and evidence-grounded when direct reputation shortcuts are suppressed. The associated code and dataset are accessible at the \href{https://anonymous.4open.science/w/FraudMessageWalker-Bench}{anonymous link}.

TimeLens is an AI-powered bilingual mobile guide for the Grand Egyptian Museum (GEM). Pointing a phone at an exhibit, a visitor sees the artifact recognized in real time and can ask follow-up questions answered in English or Arabic. The work addresses three problems specific to in-gallery deployment: fine-grained visual similarity among 51 catalogued artifacts (many near-identical Ramesside statues), the gap between curated training data and handheld camera conditions, and the risk of an AI guide stating unsupported historical facts. Two engineering contributions are reported. First, an on-device artifact detector was developed through a data-quality-driven iteration study -- from foundation-model auto-annotation (YOLO-World), through spatial label-cleaning rules, to a fully hand-annotated dataset -- isolating label quality as the decisive factor: the final YOLOv8n model resolves every previously failing class while remaining a 5.97 MB TensorFlow Lite asset that runs in real time on a mid-range phone (mAP@0.5 = 0.995, mAP@0.5:0.95 = 0.924). Second, a bilingual Retrieval-Augmented Generation (RAG) guide, grounded in a 108-record ChromaDB knowledge base, was benchmarked across seven candidate language models, with Gemma 4 E2B (Q4 K M) selected; ten targeted optimizations reduce end-to-end latency from over 30 s to approximately 10 s. Both subsystems are integrated in a production Flutter application with bilingual interface, museum location gating, and text-to-speech support.

Humor plays a central role in human social relationships, and recent advances in computational humor create new opportunities for integrating humor into human-robot interaction (HRI). While large language models (LLMs) can generate diverse forms of humor, it remains unclear how humor style, joke content, and language preference shape perceptions of robot-delivered humor in group settings. In this exploratory study, we employed a mixed factorial design in which participants evaluated AI-generated jokes delivered by a robot in a university classroom. We examined the effects of humor type (Affiliative, Self-Enhancing, Aggressive, Self-Defeating) and joke content (person-related vs. political) on perceived funniness and appropriateness, as well as preferred language. Results show that humor type significantly influences funniness, with Aggressive and Affiliative humor rated higher, while joke content primarily affects appropriateness, with person-related jokes preferred over political ones. Language preference was shaped by both joke content and participants' self-reported fluency and humor practices.

The rapid progress of large language models (LLMs) has opened up a new frontier for automatic speech recognition (ASR), making their effective integration a critical and challenging research direction. To this end, this work proposes a projector-based LLM-ASR framework targeting the key challenges of multilingual generalization and modality alignment. Our approach incorporates a Mixture of Experts (MoE) architecture to improve cross-lingual adaptability, and a Continuous Integrate-and-Fire (CIF) mechanism for dynamic downsampling and modality alignment. Experimental results show that the combination of these components yields substantial performance improvements, surpassing strong baseline models. The proposed method represents a step toward building more accurate, robust, and generalizable LLM-based ASR systems.

We investigate whether commonly available LoRA variants have an advantage over basic LoRA in multilingual instruction tuning. Experiments involving LoRA and four other variants on two datasets across diverse target languages show that there is no significant advantage in using more complex LoRA variants instead of basic LoRA, with respect to balancing cross-lingual transfer and knowledge retention. An analysis of hidden embeddings reveal that layer-wise language representation remains largely similar across LLMs fine-tuned with different LoRA techniques, suggesting that architectural novelty of LoRA techniques may not translate into better cross-lingual adaptation.