Linguistics

Automatic speech recognition for children's reading remains underdeveloped for most African languages, including Bambara, despite its potential value for reproducible literacy assessment. We present an open-source system for assessing children's reading in Bambara, developed through an end-to-end process linking field data collection, benchmark construction, model adaptation, a reading application, and classroom validation. A mobile collection and assessment app was used to collect 55 hours of raw reading speech from 60 children, from which we construct a public benchmark for Bambara child-reading assessment. Fine-tuning experiments compare Soloni, a Bambara-adapted Fast-Conformer ASR framework with TDT and CTC decoders, with QuartzNet, a compact convolutional ASR architecture. The best Soloni model reduces WER from 0.42 to 0.22 and CER from 0.15 to 0.08, substantially outperforming QuartzNet on the isolated benchmark. The experiments further show that repeated readings of the same texts provide architecture-dependent benefits: they substantially improve QuartzNet but add only marginal gains for Soloni, while SpecAugment regulates training without exceeding the best unaugmented configuration. Disaggregated analysis identifies children under 10 as the main source of residual errors, motivating targeted collection from younger readers. Ten classroom trials supported continued use of the application.

Rapid advancements in generative speech technology have compromised the reliability of voice biometrics. While current spoofing detectors excel when assessed under in-domain conditions, generalisation to out-of-domain settings is often poor. We show that this can be due to linguistic bias. A reliance on linguistic cues observed in training data can then compromise robustness to cross-data. We propose a linguistic-invariant spoofing detection framework utilizing teacher-student adversarial learning. The linguistic-aware teacher model, pre-trained on linguistic content of an external dataset, guides the student detector via gradient reversal to minimize the linguistic information. To prevent the inadvertent removal of non-linguistic cues, we incorporate a Variational Information Bottleneck to enable suppression of principal cues. Across nine DF Arena datasets, our method achieves up to a 36.2% relative reduction in the EER compare to the baseline.

WiFi sensing based on Channel State Information (CSI) promises ubiquitous, device-free perception, yet current research remains trapped in a Tower of Babel - fragmented into isolated silos where models are tailored to specific hardware dialects, fixed environments, and narrow tasks. The primary bottleneck is the Heterogeneity Gap: the disparity in signal dimensions, sampling rates, and semantic labels that prevents cross-system understanding. To bridge this gap, we propose a foundation-model framework that treats CSI not merely as raw signals but as a structured language with a learnable universal grammar. We first curate and standardize a large collection of heterogeneous real-world CSI datasets, establishing a unified infrastructure that allows incompatible signal formats to be treated as a single corpus. Second, we introduce a modular architecture that acts as a universal translator where lightweight dataset-specific adapters tokenize diverse signal inputs into a shared latent vocabulary, while a shared self-supervised Transformer backbone learns the temporal syntax of human motion and environmental dynamics. This design decouples sensing semantics from hardware syntax. Extensive evaluations show that by mastering this universal language, our approach consistently outperforms task-specific baselines and exhibits strong generalization capability in new environments, achieving superior efficiency in few-shot scenarios. By effectively absorbing heterogeneity, the framework offers a path toward robust, general-purpose wireless sensing, mirroring the linguistic generalization observed in Large Language Models. The code implementation is available at: https://github.com/cjychenjiayi/WiLLM.

Managing our emotional responses to events is key to emotional well-being, a process referred to as emotion regulation in psychology. Previous work has established that the degree to which we distance events is a type of emotion regulation. When we psychologically distance from events there can be markers in our language. These markers have been referred to as linguistic distancing. We build upon a previous metric to operationalize linguistic distancing, and explore how it changes across the lifespan. We explore this systematically by analyzing large amounts of social media text, a venue where people express their emotions. By investigating how distancing varies across age groups we can better understand how emotion regulation varies with age and provide initial benchmarks on social media data. We provide additional evidence further strengthening the hypothesis that linguistic distancing occurs in proportionally more instances with age. These findings align with past work in psychology which indicate improved well-being with older age. Better understanding how linguistic distancing changes with age is important because it functions as a marker of well-being and can inform effective health interventions. We provide a foundation for further exploring emotion regulation through linguistic distancing in text data.

Russian and Arabic are among the major languages of scientific communication. Language barriers impede the exchange of research results between these communities, which affects international collaboration and the progress of sustainability-related research. We present a benchmark for Arabic--Russian scientific translation. The benchmark includes a hybrid parallel corpus of about 27,000 sentence pairs, compiled from scientific abstracts and general-domain texts (religion, news, conversations). We fine-tune three multilingual language models -- mT5-base (580M parameters), NLLB-200-distilled-1.3B (1.3B), and Qwen2.5-7B-Instruct (7B) -- using LoRA with ranks 8, 16, 32, and 64. The Qwen2.5-7B model with QLoRA (rank 8) yields BLEU 23.15, chrF 43.89, BERTScore 0.906, and COMET 0.758. These are +4.36 BLEU and +0.051 COMET above the zero-shot baseline. Few-shot prompting with three examples does not improve performance, indicating that domain-specific fine-tuning is required. We release the models, the corpus, and the evaluation code. By lowering the language barrier for scientific texts, the work enables knowledge exchange between Arabic-speaking and Russian-speaking researchers. It contributes to sustainable partnerships (UN SDG 17) and innovation infrastructure (SDG 9), aligning with the conference's focus on technology-driven sustainable development.

This paper describes our submission to SemEval-2026 Task 9 on detecting multilingual, multicultural, and multievent online polarization. We address all three subtasks: binary polarization detection, polarization type classification, and manifestation identification for English and Swahili. Our approach leverages transformer-based models (RoBERTa-base for English, AfroXLMR-base for Swahili) with class-weighted loss functions to address severe label imbalance and per-label threshold tuning to optimize multi-label classification. On the test set, we achieve F1 macro scores of 0.7901 (English) and 0.7910 (Swahili) for Subtask 1, 0.4615 (English) and 0.4808 (Swahili) for Subtask 2 and 0.4791 (English) and 0.5830 (Swahili) for Subtask 3, which give competitive performance on the leaderboard, demonstrating the effectiveness of our methods for handling imbalanced multi-label polarization detection. Our error analysis reveals that models struggle with dehumanization detection and lack of empathy.

Recent work suggests that transformer language models show a bias towards human languages over unnatural ("impossible") languages argued to be unacquirable by humans. However, this literature has largely based these claims on differences in sample efficiency and test-set perplexity, rather than on direct evaluations of the linguistic capacities that could plausibly explain non-attestation in human languages. We evaluate two theoretically motivated linking hypotheses: impossibility arising from deficiencies in grammatical sensitivity or generative production. Using GPT-2 style models trained on perturbed "impossible" variants of English, we measure sensitivity to grammaticality using BLiMP minimal pairs, finding that model performance exhibits only gradual degradation, mediated by the language's information locality. In contrast, these models exhibited pronounced failures in generation, producing substantially fewer high-quality sentences at longer lengths. Together, these results suggest generative deficiency and transmission failures as a plausible linking hypothesis between language model behaviour and non-attestation of impossible languages.

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.

The rapid integration of Large Language Models (LLMs) has driven the evolution of Multi-Agent Systems (MAS), where specialized agents collaborate to execute complex workflows. Effective orchestration in these environments requires robust routing mechanisms to efficiently allocate tasks to the most suitable agent. However, existing routers fundamentally rely on unverified proxies, ranging from textual self-descriptions to static surrogate representations, to gauge an agent's competence. This reliance on non-empirical data creates a critical gap between an agent's projected profile and its actual operational capabilities, introducing severe security vulnerabilities. Malicious agents can easily misrepresent their proficiencies or harbor covert backdoors that evade both standard external analysis and static representation-learning techniques. In this work, we introduce ANTAP (Automatic Non-Textual Agent Picker), an evaluation-driven routing architecture that discards indirect proxies in favor of active capability testing. By dynamically querying agents to ascertain their true competencies empirically, ANTAP distills performance into fixed behavioral operators within a shared semantic space. At inference time, routing is performed via a purely non-textual algebraic projection, establishing a "linguistic firewall" that renders metadata-based attacks inexpressible. In our experiments, ANTAP achieves near-zero ASR against description-based injection attacks, compared to 67.3% and above for the description-based router baseline. Against adaptive embedding attacks, ANTAP achieves substantially lower ASR than the embedding-based baseline, with a 20% reduction, while remaining resilient to description manipulation by design.

Conversational data collected in domains such as healthcare or social sciences is a valuable resource for research and automated analysis. However, responsible data sharing requires the detection and removal of personally identifiable and sensitive information to protect individual privacy. To support the development and evaluation of automatic de-identification systems, we present DialogPII, a multilingual dataset of synthetic dialogs and speech-derived transcripts for personal information detection. DialogPII covers eight interaction scenarios (emergency calls, medical anamnesis interviews, therapy sessions, insurance communication, customer support, clinical interviews regarding an AI-supported dashboard, police reports, and group therapy discussions), 19 entity types, and 11 languages (English, Arabic, Finnish, French, German, Hindi, Italian, Polish, Portuguese, Spanish, and Turkish). Dialogs were generated semi-automatically using large language models, manually curated for plausibility and diversity, and localized to country- and city-specific contexts. All dialogs were additionally converted to speech via text-to-speech synthesis, transcribed with Whisper, and annotated through automatic projection and manual correction, yielding aligned written and speech-derived resources across all languages. We further release baseline multilingual named entity recognition models and provide technical validation through inter-annotator agreement analysis, translation quality evaluation, annotation projection assessment, and benchmark experiments with transformer-based sequence labeling models.

Safety benchmarks often buy scalability by fixing the prompt, the language, and the turn structure. For emotional-support chatbots, that bargain hides precisely where safety failures emerge: across a multilingual, multi-turn crisis conversation. We present EMPATH, a benchmark for safety evaluation of emotional-support chatbots. An auditor model role-plays help-seeking users, generating multi-turn conversations from 140 seed instructions and 34 personas. A judge model scores each full transcript against 19 metrics across five dimensions: crisis handling, therapeutic quality, conversational integrity, emotional safety, and cultural adaptation. EMPATH is built for Mexican Spanish and US English; the studies reported here run in Mexican Spanish. Auditor and judge are drawn from different model families, and the judge is treated as an instrument to be calibrated rather than trusted. A strict per-criterion rubric reveals material score inflation on 10 of the 19 metrics and restores discrimination. We study the measurement properties of the benchmark through judge calibration and cross-family inter-judge agreement. We also illustrate EMPATH on three frontier models, one of them open-weight. Aggregate scores sit within 0.74 points of one another, but per-metric profiles diverge by up to six points in model-specific places. Under the standard rubric, both the ranking and the weak spots are stable across a second, cross-family judge: 93% of scores fall within plus or minus 1. A five-run test-retest adds a second axis: even the steadiest model swings from 2 to 10 on a crisis metric across identical re-runs, and deepseek-v4-pro returns a different conversation on every run even at temperature 0. Run-to-run reliability is therefore a per-model safety property, not noise to average away. EMPATH is system-agnostic; the pipeline, seeds, personas, and rubrics are released for reuse.

Quantifying how meaning propagates through communicative exchanges remains underdeveloped in computational linguistics. Here we introduce an information-theoretic framework that quantifies the directed flow of semantic content between interlocutors and decomposes multi-source contributions into redundant, unique, and synergistic components. Our approach leverages large language models as probabilistic estimators of natural language to compute two measures: semantic transfer entropy (STE), which captures directed predictive influence between speakers, and semantic partial information decomposition (SPID), which resolves how multiple sources jointly shape a target's language. Across four experiments we show that the framework detects reduced information flow in cognitively rigid dialogue, captures the dominant role of persuaders in shaping discourse, distinguishes high- from low-quality psychotherapy by the directionality of therapist-client information exchange, and reveals synergistic premise contributions in argumentative essays. This framework opens new avenues for studying information dynamics in digital discourse, pedagogical interactions, clinical dialogues, and any domain in which the structure of linguistic exchange is of research relevance.

Human adults can often perform a novel task correctly on the first attempt after only receiving verbal or written instructions. This rapid instructed task learning (RITL) is a hallmark of human cognitive flexibility, yet its mechanisms and parallels in artificial systems remain under-explored across disciplines. In this position paper, we argue that humans possess an evolved instruction-following bias -- an inductive bias shaped by evolution to interpret and execute linguistic instructions which critically enables fast generalization of behavior from language. This bias functions analogously to the way large language models (LLMs) leverage instruction tuning to achieve zero-shot task performance. We synthesize evidence from cognitive science, neuroscience, and machine learning research to support this hypothesis. While instruction-following in AI is currently achieved via specialized training protocols, we posit that in humans it arises as an innate cognitive architecture feature. We outline testable predictions and call for more interdisciplinary research to investigate Instruction-Following as a unifying mechanism enabling rapid task learning in both natural and artificial neural networks.

When humans translate, not every word depends equally on the surrounding context. Some tokens, particularly function words like pronouns and auxiliaries, rely heavily on preceding or following sentences, while others, such as proper nouns, do not. Understanding this inherent context sensitivity is essential for evaluating whether machine translation systems use context in human-like ways. However, existing approaches to analysing context usage rely on discourse-specific test sets or model internals, making them narrow or model-dependent. We propose a post-hoc, model-agnostic framework to quantify context sensitivity at lexical and syntactic levels using two measures derived from word alignments: fertility (number of target tokens generated per source token) and entropy (stability of fertility patterns across contexts). Using reference translations for three language pairs (German $\leftrightarrow$ English, English $\rightarrow$ Hindi) under four context conditions, we show that context selectively redistributes generative responsibility from source to context tokens without altering overall fertility. Function words show the largest fertility reductions, while content words remain stable, suggesting that context resolves ambiguity rather than adding new information. Our framework provides a ground-truth characterisation of selective context usage in human translation, establishing a diagnostic baseline for evaluating machine translation models.

IndicTrans2 is the strongest open English to Indic translation system, but like most systems it is trained on general text and tends to sound stiff on casual, conversational input. We adapt IndicTrans2-1B to conversational register across all 21 Indic languages using only public data (OpenSubtitles, BPCC-H-Daily, Tatoeba). Plain fine-tuning improves conversational chrF but forgets the general domain (it drops 3.9 chrF on FLORES for Hindi). Mixing general data back into training (experience replay) and then averaging the fine-tuned weights with the base (model souping) removes that trade-off: the resulting model beats IndicTrans2-1B on conversational chrF in every one of the 21 languages (mean +6.2) while matching it on FLORES (mean change -0.17, all within 0.7 chrF). Paired bootstrap tests confirm the conversational gains are significant (p <= 0.004) and that FLORES is not significantly degraded. We are deliberate about scope: these are chrF gains, and a blind human plus multi-model LLM check does not confirm them as a perceived quality improvement, so we treat the conversational gain as largely a register match to the references rather than proof of better translation. The techniques are not new; the contribution is the honest, end-to-end study in the Indic conversational setting.

Encoders have become the state of the art for multiple NLP tasks, especially those requiring deep contextual understanding. While multilingual models offer broad coverage, dedicated monolingual encoders are essential for capturing the unique lexical and syntactic nuances of specific languages. For Portuguese, however, existing monolingual options like BERTimbau and Albertina have not kept pace with recent architectural breakthroughs, often lagging behind English benchmarks in scalability and efficiency. This work introduces BERTomelo, a next-generation monolingual encoder pre-trained from scratch and specifically optimized for the Portuguese language. By leveraging the ModernBERT architecture, BERTomelo overcomes the limitations of previous models, offering Base and Large versions with a 1,024-token context window and hardware-level optimizations like FlashAttention and alternating attention mechanisms. The model was trained on ClassiCC-PT, a massive, high-quality Portuguese corpus of 106 million documents, ensuring superior alignment with the language's contemporary usage. The results demonstrate that BERTomelo not only outperforms previous Portuguese encoders but also provides a more robust and efficient alternative to massive multilingual models in downstream tasks such as STS and NER.

Creative Commons licenses dominate African NLP corpus releases, but their compatibility rules are rarely applied. CC-BY-SA and CC-BY-NC cannot be combined in a single published dataset; a NoDerivs clause silently prohibits tokenisation and annotation. This paper audits the license provenance of over twenty corpus families used in African NLP, constructs a six-tier compatibility matrix, and applies it to three case-study languages: Kituba/Munukutuba, Zarma, and Moore. Four failure modes are documented with primary-source evidence: outright prohibition (JW300, removed from OPUS after a legal audit confirmed Terms of Service violation); composite license misrepresentation (WAXAL, whose CC-BY 4.0 claim is contradicted by its own HuggingFace dataset card); a NoDerivs clause hidden behind a CC-BY label (Tanzil); and data persistence failure (the Congolese Radio Corpus, where 402 of 405 source URLs are now dead). A pre-annotation due diligence checklist and a survey of legally clean enrichment opportunities close the paper.

Fine-tuning a large language model is a ubiquitous method for enhancing its capability on a specific downstream task. However, prior work has shown that this increase in capability comes with a cost: it can increase a model's tendency to respond to unsafe adversarial prompts, even when fine-tuning with non-adversarial data. We present the first comprehensive empirical study of this phenomenon in multilingual settings by fine-tuning Llama-3.2, Qwen3, and Gemma-3 models using benign data translated across nine languages. We find that safety outcomes are highly sensitive to both the choice of fine-tuning language and the evaluation language, with adversarial compliance rates increasing four-fold in some settings. Multilingual safety drift is decoupled from general capability metrics, and occurs heterogeneously across languages and models. Fine-tuning in non-English languages often induces smaller internal representational drifts than English, but these shifts lead models to default to either exaggerated compliance or refusal. As such, assessing fine-tuning impacts solely in English provides inadequate assurance for deployment. To facilitate further research into these cross-lingual safety blind spots, we release the Multilingual-Benign-Tune dataset and the SORRY-Bench-Multilingual evaluation suite.

Sign Languages (SLs) are the primary means of communication for millions of deaf individuals, yet existing evaluation metrics for generated SL remain simplistic and poorly aligned with human judgements. We introduce BackTranslation2.0, a linguistically grounded evaluation metric for text-to-sign translation that moves beyond naïve backtranslation. Our approach adopts an agentic framework in which a deterministic pipeline orchestrates a suite of specialised tools to assess four scoring dimensions - grammatical correctness, phonological accuracy, motion fluency, and generation fidelity - aligned with human rater assessments. Tool outputs are not treated independently: a set of large language model (LLM)-based cross-referential comparison modules evaluates consistency across tools and checks outputs against linguistic expectations, enabling structured reasoning over grammatical, phonological, and motion-level evidence. Final dimension scores are computed through deterministic weighted formulas over validated tool outputs. To validate BackTranslation2.0, we introduce and evaluate on a British Sign Language (BSL) dataset rated in a human rater study across the same quality dimensions, following a protocol developed in collaboration between linguists and deaf experts, benchmarking against six baseline metrics. Our method demonstrates strong correlation with human judgements across all dimensions, providing a more comprehensive, interpretable, and linguistically principled evaluation framework for sign language production systems.

Sign language translation (SLT) remains challenging due to its high spatio-temporal complexity, long sequences, and the need to model multiple articulators without relying on gloss annotations. Existing approaches are typically tailored to individual datasets or languages and struggle to scale, while overlooking the relationships between sign languages that could inform more effective cross-lingual transfer. We present \textbf{SIGNET}, a framework that enables motion-level knowledge transfer for cross-language sign language translation. Our key insight is that, although sign languages differ in grammar and lexicon, pretrained models capture motion-level visual patterns that can be reused across datasets and languages. \textbf{SIGNET} integrates multiple pretrained sign language backbones through an attention-based, hand-prior aggregation mechanism that guides a gated fusion network in dynamically selecting the most relevant experts. Comprehensive experiments on four benchmarks (How2Sign, Phoenix14T, CSL-Daily, and MeineDGS) demonstrate state-of-the-art translation performance, and \textbf{SIGNET} also surpasses prior methods on WLASL for sign language recognition.

We develop a quantitative theory of the Random Language Model (RLM), an ensemble of stochastic context-free grammars, in a scaling limit where the number of hidden symbols $N \to \infty$ while the grammar temperature $\tildeε_d \to 0$ at fixed $x = {\tildeε}_d \log N$. In this limit, the model admits a controlled description based on a large-deviation principle over rule-usage patterns. A semi-annealed approximation maps the problem to a class of Random Energy Models with nontrivial combinatorics. We show that the RLM exhibits a condensation transition at a critical value $x_c=1/8$, below which rule usage concentrates and language statistics acquire a nontrivial dependence on corpus length. A second characteristic scale at $x=1/2$ marks the onset of entropy reduction from its maximal value. Across these regimes, we derive explicit scaling laws for the number of distinct rules, entropy, and related observables, identifying distinct scaling, saturation, and critical regimes controlled by the interplay of grammar size, corpus length, and temperature. The theory resolves previous ambiguities regarding the existence of a thermodynamic transition and explains the slow approach to the large-$N$ limit as a consequence of the dependence on $\log N$. It further provides a unified framework in which universal statistical properties of language emerge from typical realizations of generative grammars, with implications for both natural language statistics and the behavior of large language models.

Early detection of dementia through speech analysis offers a non-invasive screening alternative, but capturing both acoustic and linguistic biomarkers remains challenging. We propose a multimodal framework leveraging Whisper for dual-purpose extraction: acoustic representations from encoder outputs and transcripts via automatic speech recognition (ASR). For the acoustic pathway, temporal networks with attention pooling aggregate variable-length sequences into fixed-dimensional embeddings. For the linguistic pathway, we prompt a large language model (LLM) to extract interpretable features spanning lexical diversity, syntactic complexity, semantic coherence, and discourse patterns. A gated fusion network integrates both modalities. On ADReSS and ADReSSo, our method achieves F1-scores of 89.47% and 90.14%, demonstrating effective integration of acoustic and LLM-augmented linguistic features. Ablation shows that multimodal fusion consistently outperforms either modality alone.

This paper describes team HSA_CORAL's submission to the FinCausal 2026 shared task on extracting cause-effect relations from financial narratives via extractive question answering in English and Spanish. We compare three modeling families: (i) encoder-only token tagging with multilingual BERT, (ii) encoder-decoder generation with multilingual BART, and (iii) decoder-only LLMs (Llama 3.1 and GPT variants) using prompt refinement, few-shot demonstrations, and supervised fine-tuning. Across settings, prompting and few-shot examples yield competitive performance, while supervised fine-tuning provides the largest gains. Our best system, GPT-4.1 Mini fine-tuned on combined English and Spanish training data, achieves a tied highest score on the English subtask (score 4.8140) and ranks third on Spanish (score 4.7753) under the shared task's LLM-as-a-judge metric. Overall, the results highlight the value of task-specific adaptation and multilingual fine-tuning for cross-lingual transfer in financial causality QA.

To avoid moderation and surveillance on social media, some users routinely invent indirect linguistic expressions (ILE) that camouflage sensitive meanings. Such expressions surface as algospeak, euphemisms, and adversarial obfuscation, depending on intent and context, and they involve recurring encoding mechanisms. We propose a comprehensive, mechanism-oriented taxonomy of ILE that abstracts away from communicative goals and instead categorizes the underlying operations through which meaning is encoded and recovered. We evaluate the taxonomy by incorporating it into LLM prompts and comparing it with four existing taxonomies and a no-taxonomy baseline, using 2,000 manually annotated TikTok and Bluesky posts. The proposed taxonomy attains the strongest document- and span-level performance across the three LLMs, achieving an improvement of 4.7% in accuracy and 5.4% in F1 over the best-performing benchmark. The empirical results reveal the importance of a comprehensive, mechanism-oriented taxonomy as a stable scaffold for detecting emerging coded language and a useful input to content moderation. Disclaimer: This paper contains content that may be profane, vulgar, or offensive.

Translation cascades for reasoning translate the query from another language to English, reason in English, and translate the answer back to the original language. This is a competitive approach to multilingual reasoning, but structurally lossy, since each stage discards information later stages may need, including cues for cultural grounding, register, and disambiguation. We examine the benefits of a simple and training-free intervention: a context-aware translation cascade, which additionally provides the original question, the English translated question, and the reasoning trace to the context of the final translation module. We evaluate gains across nine multilingual benchmarks including various task types, three backbone models, and 285 high-, mid-, and low-resource languages, and demonstrate strong gains for open-ended generation across models and resource regimes. We show that the original language question carries most of the beneficial context. Our study emphasizes the need to better design information flow in machine translation cascades for mitigating error propagation, and provides a simple and actionable default strategy: preserve the original user question until the end of the pipeline.

Garden path sentences present a processing difficulty for humans -- the sentence prefix leads the listener towards one interpretation, until the listener hears a critical word that shows that the initial interpretation was wrong. Lexical surprisal, a measure that usually predicts sentence processing difficulty quite well, fails to provide good predictions for garden path sentences. We propose an alternative that actively predicts a probability distribution over syntactic trees (its syntactic belief) and updates that distribution after each new word. If a processor is led down a garden path, syntactic beliefs will be wrong and will require a large update at the critical word. The magnitude of the update is measured with a generalized Rényi divergence. Crucially, this metric is dependent on lexical items, but is fully independent of the probability of lexical items. This Syntactic Belief Update provides a better fit to the human reading time data on garden path sentences. This suggests a new research direction examining purely non-lexical alternatives to surprisal for psycholinguistics.

Speech-capable models are increasingly deployed in real-world applications across languages. Yet their safety and fairness beyond English settings and under naturalistic conditions remain understudied. We survey safety reporting practices across state-of-the-art speech model releases, finding that only 8% document any multilingual analysis. To address this gap, we introduce RedVox, a multilingual safety and fairness benchmark for audio and speech built on real voices, covering unsafe and unfair stereotypical requests across five languages (English, French, Italian, Spanish, and German). Evaluating eight state-of-the-art models, we find that vulnerabilities persist even under non-adversarial conditions, worsen in non-English languages, and are amplified when the request comes from a spoken input. Finally, by surveying the participants who contributed to RedVox, we document the unique personal and privacy challenges of collecting speech data with human participants, pointing to broader sociotechnical challenges in naturalistic speech safety research.

Concept Bottleneck Models (CBMs) promise transparent reasoning by predicting through human-interpretable concepts, yet their effectiveness fundamentally depends on how well visual and textual representations are aligned or matched. Existing vision-language CBMs often rely on pre-aligned encoders or global cosine similarity, which obscures fine-grained concept localization and fails to reflect true semantic geometry. In this work, we rethink concept alignment as a dynamic cross-modal transport process instead of static projection and propose the Optimal Transport Flow Concept Bottleneck Model (OTF-CBM). It first learns a data-driven semantic cost via Inverse Optimal Transport to measure cross-modal distances, and then performs unbalanced optimal-transport-based flow matching to model semantic transitions between visual patches and textual concepts. With velocity-based concept activation, OTF-CBM captures interpretable geometric relations without ODE integration. Experiments further show that OTF-CBM achieves superior classification accuracy and concept faithfulness, offering a new geometric and dynamical perspective for interpretable cross-modal reasoning.

In today's fast-paced information era, logical fallacies, defined as defective patterns of reasoning, inevitably contribute to the growth of information disorder. However, often fallacies appear in nuanced forms that complicate automated classification. In this study, we investigate whether merging abstract logical structures with context-level linguistic cues proves beneficial for fallacy classification, developing a framework that inductively extracts such patterns from fallacious examples and their explanations using Large Language Models (LLMs). We evaluate the impact of these patterns across different LLMs and experimental zero- and one-shot configurations, showing statistically significant improvements over zero-shot baselines and outperforming competing approaches. Cross-dataset experiments validate generalization, establishing data-driven pattern extraction as an effective method for generating logical representations.

Multilingual large language models often produce inconsistent reasoning and answers for semantically equivalent prompts in different languages. Prior work suggests that intermediate representations can be relatively language-agnostic, but generation becomes increasingly language-specific as models commit to discrete output tokens. This is problematic because language-specific lexical choices can cause semantically equivalent reasoning paths to diverge across languages. These divergences motivate searching for a cross-lingual alignment signal that is less tied to any single vocabulary item or script. We propose SOLAR, an auxiliary objective for supervised fine-tuning that aligns soft-token representations across languages, using English as a pivot. Soft tokens are probability-weighted mixtures over the vocabulary embeddings, yielding continuous representations that can aggregate information from semantically related tokens across languages. We then align each non-English soft-token summary to its English counterpart in the shared embedding space. Across four multilingual reasoning benchmarks, SOLAR improves accuracy by up to +17.7 points over the base model and +3.8 over standard supervised fine-tuning, with the largest gains on low-resource languages. SOLAR also strengthens final-layer cross-lingual similarity and substantially reduces language-cluster separability, suggesting that aligning soft-token representations helps preserve shared semantic structure during multilingual reasoning.

The neutral, or floating, tone of Mandarin Chinese is a tone with an enigmatic set of properties. It has been described as a reduced tone, or as a tone that sometimes is lexically fixed but that can also be toneless. In two-syllable words, it is found only on the second syllable, but single-syllable words can also have the neutral tone. We present a corpus-based study of the phonetic realization of the neutral tone in spontaneous conversational speech corpora of Beijing Mandarin and Taiwan Mandarin. We show that the neutral tone has its own tonal target, just as the four lexical tones of Mandarin. We also show that disyllabic words with a neutral tone have pitch contours that have a pitch component that depends on the tone on the first syllable, just as has been observed for two-syllable words with a lexical tone on the second syllable (Chuang et al., 2026). Furthermore, words with a floating tone have word-specific pitch signatures, which have also been documented for single-syllable words (Jin et al., 2026) as well as two-syllable words (Lu et al., 2026b). These word-specific pitch signatures are shown to be predictable to some extent from words' contextualized embeddings, as previously reported for lexical tones (Chuang et al., 2026; Lu et al., 2026b). As there is also considerable variability in the realization of lexical tones, we propose that the neutral tone is, in fact, a lexical tone in both Taiwan Mandarin and Beijing Mandarin. We document both similarities and differences in the realization of the floating tone in these two varieties and provide evidence, using contextualized embeddings, that some of the observed differences may arise from differences in the meanings of the words as used in the two corpora.

Text detoxification, the automated detection and mitigation of abusive and harmful content, is essential for ensuring the safety of online communities and protecting users. However, low resource languages such as Tatar have received little research attention. In this paper we present Tatoxa, a novel state-of-the-art system for text detoxification in the Tatar language. Comparative experiments show that the proposed approach outperforms existing open source and proprietary commercial LLMs on key quality metrics. We also introduce a new dataset for text detoxification in Tatar, designed for fine tuning and evaluation in low resource settings. Finally, cross lingual transfer experiments indicate that transfer from other languages, including the culturally close Russian, performs significantly worse than training on native Tatar data even when a large Russian corpus is available.

Sparse Mixture-of-Experts (MoE) architectures have emerged as an increasingly influential paradigm as they offer a strategic balance between parameter scalability and computational efficiency. However, low-resource languages, which suffer from a scarcity of high-quality training data, often have their tokens routed to different experts than those predominantly activated by high-resource inputs, which limits cross-lingual expert sharing. This cross-lingual routing divergence consequently hinders their efficacy in multilingual contexts. To address this issue, we propose SARA (Semantically Anchored Routing Alignment), a framework designed to transfer specialized capabilities from high-resource languages as anchors to low-resource languages. SARA explicitly aligns the routing distribution of multilingual inputs with high-resource semantic anchors using a symmetric Jensen-Shannon (JS) divergence constraint. Unlike traditional distillation methods that operate on output logits, SARA directly aligns the internal routing distributions of MoE layers, encouraging mechanistic consistency in expert selection across languages. We conduct experiments on 2 LLMs across 5 low-resource languages and 3 benchmarks. Experiment results demonstrate that SARA outperforms standard instruction tuning, e.g., +0.8% on Qwen3-30B-A3B and +1.2% on Phi-3.5-MoE-instruct on Global-MMLU. Further analyses show that SARA effectively addresses performance bottlenecks in low-resource languages, providing a scalable pathway to enhance multilingual capabilities in sparse architectures.

Recent LLMs demonstrate strong mathematical reasoning capabilities, but existing gains rely heavily on English-centric training resources and benchmarks. As a result, reasoning performance degrades substantially in low-resource languages such as Urdu, where reasoning-oriented datasets and adapted models remain scarce. Urdu lacks both reasoning-oriented resources and models adapted for multi-step mathematical problem solving, limiting the applicability of recent progress to Urdu-speaking users. We address this gap through Riazi-8B, an Urdu mathematical reasoning model developed through a two-step adaptation process comprising continued pre-training on Urdu Wikipedia and supervised fine-tuning on Urdu Chain-of-Thought data derived from GSM8K. We evaluate Riazi-8B on MGSM-Urdu against existing Urdu instruction-tuned models. Our results show consistent improvements in answer correctness, reasoning quality, response completeness, and Urdu generation. Our findings demonstrate that combining Urdu language adaptation with reasoning-focused fine-tuning is an effective strategy for extending mathematical reasoning capabilities to low-resource languages.

Recent advances in sequence modeling have significantly improved ASR systems, bringing them close to human-level recognition accuracy and enhancing robustness across diverse acoustic conditions and languages. In contrast, Forced Alignment has not experienced comparable progress, and traditional HMM-GMM frameworks remain widely adopted and highly competitive. To address this gap, we propose an end-to-end, fully differentiable neural architecture specifically designed for phoneme alignment. The model consists of an encoder that processes the input signal and a decoder that produces alignment decisions. The encoder is structured into two complementary branches: one dedicated to phoneme identity verification and the other to phoneme boundary detection. The decoder is implemented as a trainable module based on differentiable soft dynamic programming. The entire system is optimized end-to-end using a novel contrastive loss that encourages clear separation between steady-state phoneme regions and transition boundaries. The proposed approach outperforms the current state of the art in phoneme alignment on hand-annotated English benchmarks, achieves strong word-level generalization results, and demonstrates generalization on unseen languages.

Dialogue systems based on large language models (LLMs) have advanced significantly in recent years. However, dialectal variation remains a major challenge, particularly for systems that process spoken input. LLM-based speech language models (SLMs), which integrate LLMs with speech processing components, show promise for spoken language tasks, yet their ability to comprehend dialects has not been sufficiently studied. Moreover, it remains unclear how the dialectal understanding of the base LLM affects SLM performance. This study investigates the dialectal robustness of both LLMs and SLMs using Japanese dialects as a test case. We define robustness as the ratio of performance on dialectal versus standard inputs, enabling fair comparisons. Our experiments show that SLM robustness correlates with that of their text-based counterparts. Furthermore, training with dialectal data and fine-tuning the speech encoder each improves robustness in SLMs.

Large language models (LLMs) are increasingly deployed across languages, but their safety behavior remains uneven across linguistic and cultural contexts. This survey synthesizes work on toxicity detection and detoxification for multilingual LLMs. We first catalogue threat models that exploit language choice, translation pivots, code-switching, orthographic variation, multi-turn interaction, and post-deployment fine-tuning to weaken safety alignment. We then organize task formulations (toxic-to-neutral rewriting, toxicity classification, and toxic-generation evaluation), multilingual detection approaches (cross-lingual encoders, translation pipelines, representation-level probes, and LLM-based detectors), and mitigation strategies spanning data filtering, supervised and preference-based tuning, decoding-time steering, representation editing, and multilingual guardrails. Across these areas, we identify persistent challenges: uneven language coverage, culturally contingent definitions of harm, fragmented evaluation protocols, and the risk that detoxification suppresses legitimate dialectal or identity-related expression.

We present a study on low-resource machine translation for the Tangkhul-English (nmf-en) language pair. Tangkhul is a severely under-resourced Tibeto-Burman language spoken primarily in Manipur, India, with virtually no prior natural language processing infrastructure. We describe two systems: (1) a primary system based on ByT5-large fine-tuned on 38,336 Tangkhul-English parallel sentence pairs, and (2) a contrastive system based on mT5-small fine-tuned on the same corpus. Our primary ByT5-large system achieves a corpus BLEU score of 39.97, chrF++ of 58.07, BERTScore F1 of 0.8104, and COMET (wmt22-comet-da) of 0.7302 on a held-out test set of 3,856 sentences. We further discuss the orthographic challenges specific to Tangkhul's Latin-script diacritics, the domain bias of our training corpus (which comprises biblical text, stories, and conversational data), and avenues for future improvement through data diversification and domain adaptation.

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.

High-quality annotations of structured representations are expensive to collect over large corpora. Manual annotation of structure is laborious, and model-based annotation, although cheaper to generate, requires expensive validation and potentially significant supervision to ensure that the annotation quality is strong enough to be useful downstream. In traditional annotation workflows, annotation of each complete example is performed end-to-end by a single annotator. However, structured annotation is complex, and each aspect of the task represents a unique challenge with an associated inferential load for a given annotator. Modern annotation projects can incorporate heterogeneous groups of annotators, including both models and human annotators with varying domain and linguistic expertise. It remains unclear, however, how to redesign annotation tasks in this setting, where efforts are discriminately allocated across heterogeneous annotators with respect to distinct annotation challenges. We propose to decompose annotation tasks into sub-tasks in order to reduce the aggregate inferential load of annotation projects. Inspired by the notion of centers from centering theory, we introduce a formal model of inferential load based on the degrees of freedom in the space of valid annotations. Using this model, we show that identifying these centers (i.e. salient anchor entities realized by annotation sub-tasks) constrains the output space complexity, and decompositions which isolate and advance center identification reduce the aggregate inferential load. We provide guidelines for decomposing complex structured annotation tasks, supported by examples demonstrating improved cost-efficiency from our prior work. Finally, we present a procedure for allocating sub-tasks across annotators to maximize quality under a fixed budget.

Parametric knowledge in Large Language Models is not equally accessible across languages. As a result, standard inference techniques often struggle to surface localized facts, leading to failures in cross-lingual knowledge transfer and consistency. In this work, we investigate techniques for accessing hidden factual knowledge by exploring cross-lingual prompting strategies. We identify four inherent dimensions of cross-lingual exploration that directly govern parametric knowledge retrieval and evaluate them on multilingual factual benchmarks covering 17 typologically diverse languages. Our results demonstrate that cross-lingual exploration significantly improves knowledge transfer and factual recall, representing a more efficient compute Pareto frontier than native-language scaling. Furthermore, we observe corresponding improvements in cross-lingual consistency, exceeding what can be explained by accuracy gains alone. Overall, our work establishes multilingual prompt exploration as a highly effective inference-time strategy for unlocking latent parametric knowledge.

Scam phone calls exploit vulnerable communities worldwide, yet research on detection has focused almost exclusively on English and other high-resource languages. In low-resource settings such as Turkish, detection is especially difficult, as annotated data is scarce and technological defenses remain limited. This research investigates how large language models (LLMs) can support scam detection in Turkish by introducing the first public multi-modal dataset of 100 aligned audio-transcript pairs of scam and benign conversations. We evaluate seven LLMs spanning three model families: Gemini 2.5 (Flash, Flash-Lite, Pro), GPT-4o, and Qwen (Max, Plus, Turbo), under three input conditions: raw audio, automatic speech-to-text transcripts, and transcripts refined by a native speaker. Our results suggest that transcript-based inputs consistently outperform direct audio processing, while human-corrected and uncorrected transcripts perform comparably. By centering a low-resource language and real world threat, this work highlights the urgent need for culturally and linguistically inclusive AI safety research and more robust multi-modal systems for fraud prevention.

Commercial large language models bill, scale latency, and budget context per token. Yet tokenizers assign more subword tokens to the same meaning in some languages than in others, so speakers of languages with high token-fertility pay a structural penalty before a model is ever invoked. This penalty is documented for multilingual settings in general, but it has not been measured systematically for African languages at the level of enterprise deployment economics and cognitive context capacity. We measure it across 20 African languages spanning five language families and three scripts (Latin, Ge'ez/Ethiopic, N'Ko; 19 appear in the primary FLORES-200+ corpus, with Nigerian Pidgin measured via MAFAND-MT only), using parallel corpora so that the language effect is isolated from content. Across 11 frontier and open tokenizers on FLORES-200+, every African language carries a tokenization premium above English (median 1.88x on GPT-5 / o200k_base, up to 8.92x for N'Ko); the penalty is largest for Ethiopic and N'Ko scripts (reaching 7-9x) and is near-invariant across corpora (FLORES vs SIB-200 Pearson r = 0.9998). Translated into deployment terms, this results in up to 8.9x inference cost and an equivalent generation-latency multiplier (N'Ko vs English on GPT-5; 7.4x for Amharic), and as little as 11% of English's effective context window. The best currently available tokenizer for African languages, Gemma 4, reduces the mean premium from 3.31x (cl100k_base) to 2.38x, but no tokenizer eliminates the penalty. We release an open measurement tool (afri-fertility), a public leaderboard, a results dataset, and mitigation guidance for African builders. The penalty falls hardest on the languages whose speakers can least afford it, a digital divide encoded directly into the subword vocabulary.

The Prague Dependency Treebank framework is unique in its attempt to systematically include and link different layers of language, including a meaning representation with several types of inter-sentential phenomena, especially coreference and discourse relations. We present its second consolidated version (PDT-C 2.0), which concludes almost 30-years long project of sustained development of the resource to a uniformly and coherently annotated, genre-diversified, almost 4 million token language resource of Czech language, with accompanying fully compatible lexicons. In addition to continuous linguistic research, the richly linguistically annotated corpus is also widely used in international comparisons of the development of traditional and novel NLP tools as well as in conversions into other formalisms. The corpus and the trained parsers are available under the CC BY-NC-SA licence.

Retrieval-augmented generation (RAG) in clinical settings increasingly requires multilingual retrieval against predominantly English evidence corpora. Multilingual medical retrieval demands three capabilities: cross-lingual alignment, concept discrimination, and evidence retrieval. However, existing benchmarks evaluate these only in isolation, leaving the interaction between biomedical expertise and multilingual coverage unmeasured. We introduce MMed-Bench-IR, a benchmark designed to disentangle these axes across 6 languages and three structurally heterogeneous tasks: (1) cross-lingual medical QA retrieval with 6,127 queries grounded in the Unified Medical Language System (UMLS), (2) concept discrimination over 4,975 confusion sets at three difficulty tiers, and (3) multilingual evidence retrieval for RAG with 2,040 quality-assured queries. The three tasks share zero concept and query overlap by design, ensuring that aggregate scores reflect genuine capability breadth. Evaluation of ten systems across six paradigm families reveals severe cross-lingual failure: biomedical encoders that score 0.818 nDCG@10 in English drop to 0.056 in Japanese, a gap that English-only benchmarks cannot detect.

More than a billion people communicate in Indic languages, yet the natural language processing infrastructure serving them remains fragmented and underdeveloped. The cause is structural: the field organizes its tools and benchmarks around individual languages or small subsets of genealogical language families, building separate analyzers, parsers, and datasets for each language and starting over for the next. This overlooks a deep regularity. Through more than two millennia of convergence around Sanskrit, Indic languages came to share a morphosyntactic architecture formalized in Pānini's grammar, the Astādhyāyī. This cuts across genealogical lines, uniting languages through a common framework. We argue that this Pāninian framework supplies a unifying computational architecture the field has lacked, and that benchmarks grounded explicitly in it would make Indic language systems more accurate, more data-efficient, and more transferable, effectively merging many apparently disparate and sparse Indic language resources into a single high-resource metalanguage bedrock. We propose a four-part benchmark suite to render this shared architecture explicit, measurable, and ready to be leveraged for practical applications. Moreover, we underscore the question it raises for interpretability research: whether neural models trained on these languages come to represent Pānini's categories on their own.

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 ask whether the geographic origin of Tang-dynasty poets leaves a detectable linguistic trace in their work. Aggregating every poem attributed to each author in the Complete Tang Poems (Quan Tang Shi) and linking poets to their administrative circuit of origin via the China Biographical Database (CBDB), we build a poet-level corpus of 357 poets across the ten Tang circuits and frame origin prediction as multi-class classification. Using character $n$-gram TF-IDF together with interpretable domain features (imagery, season, and allusion), classical and neural models predict a poet's broad region (South vs.\ North) at $0.69$ accuracy, well above the $0.53$ majority baseline, and finer circuit-level origin above chance. Beyond classification, three findings emerge. (i) Linguistic distance between circuits grows with geographic distance (Mantel $r=0.40$, $p\approx0.09$ over nine circuits), evidence of a distance-decay effect in poetic language. (ii) The signal interacts with time: South/North separability is at chance in the High Tang and strongest in the Late Tang, consistent with court-driven homogenization at the empire's height followed by regional divergence. (iii) The model's confident errors are historically meaningful -- in the Early Tang, every misclassification is a southern poet read as northern, reflecting the prestige of the northern court idiom. We further show that, when given the whole corpus through a hierarchical frozen-encoder representation, a classical-Chinese transformer (GuwenBERT) only matches -- not beats -- simple TF-IDF, and that combining them adds nothing, indicating that character $n$-grams already capture the regional signal. Our results position interpretable machine learning as a hypothesis generator for literary history.

"Talk short. Drop grammar. Save token." This caveman style is widely promoted as a way to cut inference cost, but whether it actually saves anything depends on which channel (the user's prompt or the model's response) is being compressed. We present Cavewoman, a two-channel evaluation protocol that scores every generation on task accuracy, realized per-item cost, and reference-text agreement against the model's unconstrained reference. We evaluate eight models on five datasets at five reduction levels, with both channels measured on the same items. Output compression cuts realized cost on most API models (1.4-2.4x per model, up to 3x in the best case) and on all four open-weight models under public-tier pricing. Input compression has the opposite effect, a strict lose-lose: it raises net cost rather than lowering it (~1.15x on the five-benchmark mean, up to 1.8x on the worst dataset and 2.7x under stronger compression), because models compensate with longer responses even as accuracy collapses. Under the same setting, surface text diverges from the unconstrained reference: on the non-reasoning models, roughly half of all generations are correct yet their surface text no longer entails the model's own unconstrained baseline generation. The divergence survives length-controlled re-scoring, multiple-comparisons correction, and replication under complementary semantic measures. Code and data are available at https://github.com/danielle34/cavewoman.

Self-supervised and supervised speech models are increasingly used to investigate which linguistic information their internal representations encode, and at what level of abstraction they encode it. One underexplored phenomenon is consonant cluster reduction (CCR) in African American English (AAE), a widespread phonological process and a source of automatic speech recognition (ASR) disparity. To examine how CCR is represented, we conduct speaker-independent layer-wise probing of wav2vec2-base and Whisper-small using two tasks: segmental reduction detection and segmental restoration of underlying cluster identity. Both models distinguish reduced and canonical forms with high accuracy. Crucially, reduced segments retain cues to their underlying stops, indicating that CCR is encoded as structured gradient phonological variation rather than simple segmental deletion. These results demonstrate structured phonological encoding of AAE CCR patterns in modern speech models.

Tokenization is a foundational step in NLP pipelines, yet standard evaluation metrics such as fertility rate fail to capture morphological correctness for agglutinative languages. We present QuechuaTok, a systematic benchmark comparing four tokenization strategies - BPE, Unigram LM, WordPiece, and a morphology-aware PRPE tokenizer - for Southern Quechua (quz), a low-resource agglutinative language spoken by 8-10 million people in South America. Using a 200k-sentence corpus and the SQUOIA finite-state morphological analyzer (Rios, 2016) as silver standard, we evaluate three metrics: fertility rate, OOV rate, and morphological boundary accuracy (MorphAcc). Our results show that BPE achieves the lowest fertility rate (1.636 at 16k vocab) by memorizing surface word forms, while achieving only 6.67% MorphAcc. PRPE achieves 83.33% MorphAcc - the highest of all systems - demonstrating that fertility rate alone is insufficient to evaluate tokenizers for agglutinative languages. All code and models are publicly available at kaggle.com/code/macmaky/quechuatok

Language-specific tokenizers improve tokenization quality and the downstream performance of models on those languages. However, using such a tokenizer comes at a cost: either a new model must be trained from scratch, or the vocabulary of an existing pretrained model must be adapted. We propose Language-adaptive Maximum a Posteriori (LangMAP) Tokenization, a tokenization scheme that extends the UnigramLM algorithm to the multilingual setting, producing language-specific tokenization from a single shared vocabulary. Notably, LangMAP can be used when training a multilingual language model from scratch or to adapt a pretrained model's tokenizer to individual languages without changing its vocabulary. While language labels are required at training time, a key feature of the algorithm is that it then performs language-specific tokenization at inference without knowledge of the input's language. Across 14 open-source tokenizers, 9 natural languages, and 9 programming languages, LangMAP improves morphological boundary alignment and, for all coding languages tested, alignment with abstract syntax tree (AST) leaf boundaries. In fine-tuning experiments, results are mixed: LangMAP improves target-language grammatical acceptability (MultiBLiMP) on the languages tested; its benefits are less consistent on knowledge-related tasks (Global-PIQA, Belebele).

While the wider applicability of LLMs in the legal field is currently debated due to their reliability and the gravity of any errors, narrow uses with well-understood and mitigated risks have emerged. Notably the Swiss Federal Supreme Court uses small on-premises models for tentative translations and short-passage summarization across the four official languages. However, such usage is challenging in the context of Criminal Law. Since rulings and cases employees work on routinely can contain detailed descriptions of violent and sexual offenses, their legitimate work is compromised by refusals and disclaimers due to the activation of model guardrails (over-alignment). To measure this phenomenon, we introduce TF-RefusalBench, a multilingual benchmark for criminal-law translation and summarization derived from public Swiss Supreme Court rulings. TF-RefusalBench contains 5,200 total prompts across French, German, Italian, and English, corresponding to common task prompts and passages likely to trigger refusal. We then use TF-RefusalBench to show that over-alignment is a multifaceted phenomenon, influenced by the model and the prompt and text languages being processed, and that its impact cannot be evaluated solely from an over-refusal perspective, given the disclaimer's impact on task faithfulness. Finally, we evaluate approaches to enable on-premises LLMs for Criminal Law Tasks, demonstrating that while prompting can be effective, abliteration (refusal directions ablation) eliminates refusal with minimal impact on task performance.

We study the limits of CTC-internal scoring for N-best hypothesis selection and locate the information bottleneck separating acoustic confidence from linguistic plausibility. Eleven CTC-internal and acoustic-feature scoring strategies produce no statistically significant WER improvement over greedy decoding on LibriSpeech dev-other at G=16 (all p > 0.05). The exhaustion is systematic: CTC's Spearman $ρ$ between hypothesis score and per-utterance WER degrades from -0.574 at G=4 to -0.270 at G=128, a 53% loss driven by blank-path proliferation. This establishes that the discriminative capacity of CTC-internal representations is saturated: no recombination of acoustic signals can close the oracle gap. Confirming that the bottleneck is linguistic, not acoustic, external linguistic information introduced via MBR decoding breaks through it. MBR-CER decoding with a RoBERTa pseudo-log-likelihood (PLL) posterior ($τ$=10, G=128) achieves 5.42% WER on held-out LibriSpeech test-other (greedy 5.96%, $Δ$=-0.535 pp, p<0.0001, 9.0% relative). RoBERTa PLL $ρ$ degrades only 21% over the same range, retaining discriminating power where CTC loses it. Applied without retuning across two Zipformer architectures, three domains (LibriSpeech, TED-LIUM 3, VoxPopuli), and four MUSAN noise levels, the recipe gives significant gains in 11 of 13 conditions. On the training side, standard MWER training via the CTC forward-backward algorithm implements Rao-Blackwellized REINFORCE at the output projection (variance about 3x below Viterbi). Yet sequence-level fine-tuning fails at near-converged checkpoints: all four MWER configurations on CR-CTC collapse (+6.18 to +8.90 pp WER), as a training oracle gap of 0.007 pp provides no usable reward signal.

Large language models (LLMs) increasingly mediate how people form impressions of organisations, yet most monitoring is done in English, assuming an English query returns a representative picture. We measure how far that holds. We queried three grounded LLMs (GPT-5.4, Gemini 3.1 Pro, Perplexity Sonar Pro) about 66 brands from eleven Northern, Baltic, and Central European markets, in twelve languages across four families (Germanic, Uralic, Baltic, Slavic), generating 35,640 responses. Multilingual embeddings (BGE-M3) allow cross-language comparison without translation. Three results emerge. First, AI-constructed reputation is language-bound: mean cross-language cosine similarity is 0.825, same-family responses are more similar than cross-family (0.844 vs 0.820; d = 0.31), and sentiment varies by language (F = 268.5, eta^2 = 0.077), with Uralic and Baltic languages most positive and Germanic, including English, most critical; clustering recovers the Slavic and Baltic families (cophenetic 0.915). Second, query language shifts which brands are recommended far more than how they are described: moving from an English query to a brand's home language raises recommendation share by 0.80 for local champions but only 0.15 for global multinationals (t = -8.84, p < 0.001), with no comparable reversal in sentiment. An English-only audit therefore understates a local champion's AI visibility. Third, response stability varies more with model choice than with language (eta^2_model = 0.32 vs eta^2_language = 0.01, on a five-iteration replication over a 20-brand subset). These results indicate that English-only AI reputation monitoring leaves a measurable language blind spot, concentrated in the visibility of locally headquartered brands.

Optical Character Recognition (OCR) for low-resource languages is often constrained by the lack of annotated training data and the complexity of script-specific rendering. Kashmiri, written primarily in the Perso-Arabic Nastaliq script, presents additional challenges due to contextual glyph shaping, dense ligatures, and orthographic variability. We introduce Koshur Pixel, the first large-scale synthetic OCR dataset for Kashmiri, comprising 613,078 image-text pairs generated from the KS-PRET-5M corpus using the SynthOCR-Gen framework. The dataset spans multiple fonts and textual granularities, ranging from individual words to full-page documents, and incorporates more than 25 augmentation strategies that emulate real-world document degradations. Koshur Pixel provides a scalable and cost-effective alternative to manual annotation, establishing a foundational resource for training OCR systems, digitizing Kashmiri textual heritage, and advancing language technologies for a severely under-resourced language.

As Large Language Models (LLMs) achieve widespread integration across diverse linguistic landscapes, ensuring their safety and alignment with regional normative values remains a critical challenge. Current safety mechanisms are predominantly optimized for English-centric frameworks, often failing to capture the unique socio-cultural sensitivities and localized categories of harm inherent to the Indic region. To address this gap, we introduce IndicGuard, a multilingual safety guard model and dataset for Indic languages. We construct a high-volume, culturally nuanced safety dataset encompassing ten major Indic languages, systematically curated to capture regional harms, sensitive socio-political contexts, and adversarial jailbreaks. Leveraging this corpus, we fine-tune a 4B-parameter instruction-tuned model based on Gemma-3-4B-IT to serve as a multilingual safety guardrail for real-time content moderation and policy compliance checking. Our empirical evaluations demonstrate that IndicGuard significantly enhances LLM robustness against localized vulnerabilities, achieving high moderation consistency across different conversational turns. Crucially, IndicGuard consistently outperforms the existing baseline model, CultureGuard, across evaluated languages. Finally, we demonstrate that our model effectively generalizes to low-resource Indic languages excluded from training, substantiating the structural robustness and cross-lingual transfer capabilities of the framework.

This study investigates sentiment polarity biases, specifically, differences in how accurately AI models classify positive versus negative reviews across languages and model architectures. Large language models show a negative bias in French and are more accurate on negative reviews, while encoder models exhibit positive bias in Japanese, missing negative reviews that use indirect criticism. These language-specific polarity biases have implications in both social and business domains deploying multilingual sentiment analysis systems.

Multilingual Language Models like mBERT are widely used for low-resource NLP, yet their adaptation to morphologically inconsistent languages such as Roman Urdu remains underexplored. Roman Urdu spelling variation causes severe sub-word fragmentation, averaging 1.50 sub-words per token. We propose \textit{ROMEVA} (Roman Urdu Embedding-preserving Vocabulary Adaptation), which combines sub-word-average initialization and a PCA-guided anchor loss to stabilize embeddings during vocabulary expansion. Using a 36,130-comment Roman Urdu corpus, we add 500 highly fragmented tokens to mBERT and compare naive fine-tuning, sub-word-aware fine-tuning, and \textit{ROMEVA}. While \textit{ROMEVA} most effectively preserves the pretrained embedding space, naive fine-tuning achieves the strongest downstream sentiment classification performance. These findings reveal a disconnect between embedding stability and downstream performance, suggesting that stronger adaptation may be preferable to strict embedding preservation in morphologically inconsistent languages.

Why do multilingual language models sometimes generate in the wrong language, and why is this so hard to fix? We introduce Language Identity Head Ablation (LIHA), a causal intervention that zeros each attention head individually and measures the resulting language switch rate across a parallel dataset of 2,700 prompt-language pairs spanning seven languages. Applied to GPT-2, LIHA identifies a small set of first-token broadcaster heads - led by L6H1 (switch rate 0.32, 3.23 $σ$ above the population mean) - that attend persistently to the first prompt token, propagating its language signal throughout generation. Compensatory redistribution when heads are ablated is statistically significant (p < $10^{-5}$) and follows a directional, hierarchical pattern: compensation always recruits heads in layers above the ablated head, suggesting a feedforward cascade rather than global diffusion. To probe how training regime shapes these circuits, we apply LIHA to a controlled pair - Qwen2.5-1.5B-Base and Qwen2.5-1.5B-Instruct - identical in architecture and size, differing only in training. The base model is nearly flat (max SR=0.016, 200/336 heads at SR=0.0); the instruct model concentrates causal influence sharply at layer 0, led by L0H5 (SR=0.224, 8.93 $σ$ above mean), with all other layers near zero. This controlled comparison provides direct causal evidence that instruction tuning reorganizes language identity circuits toward early-layer localization. Extended experiments with Chinese and Russian confirm that first-token broadcasting is script-specific in GPT-2, with non-Latin languages handled at layer 0 - the same locus as the instruction-tuned model. Code and data will be released upon publication.