machine translation engineers

Abstract: Safety alignment for Large Language Models is heavily focused on English with limited multilingual representation, particularly damaging low-resource language performance. Related open training datasets are often limited in size or do not follow region-localized harmful topics to broaden the scope of alignment beyond cultures. We propose a scalable template-based augmentation pipeline to produce massive multilingual refusal and benign corpora, extracting safe templates from harmful prompts, translating them and then filling in the slots with culturally authentic entities, both harmful and benign. As a result, we release MSafeTemp, a validated corpus spanning 60 languages useful for multilingual safety alignment and evaluation. Using the dataset, we show that English-only evaluation misjudges multilingual safety: for several models, attacks that fail in English succeed once translated, while culturally localized variants expose further failures that translation alone misses. We also confirm that augmenting safety training data through templates allows for higher diversity and quality of alignment, resulting in improved safety performance beyond English. MSafeTemp and the respective pipeline are aimed at facilitating research and development of safer and more responsible models for less-represented cultures.

Abstract: Social media enables cross-lingual communication, yet the meaning of user-generated content (UGC) is often distributed across text, images, and cultural knowledge, making translation challenging. Existing evaluations typically treat images as optional and assume textual information is sufficient, obscuring whether errors stem from missing visual evidence or inadequate cultural adaptation. We introduce \textbf{Culture-MMT}, a diagnostic benchmark for \textbf{distributed-evidence pragmatic reconstruction} in social-media translation. It contains 498 curated, de-identified UGC instances across 20 domains, categorized into Symbol, Visual, and Hybrid by their required evidence. We also develop a human-aligned domain-specific judge, Judger, and a controlled evaluation framework that systematically varies the evidence available to translators. Across 17 model configurations, real images improve performance by +0.280 on Visual and +0.197 on Hybrid, but only +0.050 on Symbol, while random images provide no comparable gains. Human-authored descriptions further show that jointly dependent cases remain more challenging even when visual facts are explicit. These results characterize culturally effective social-media translation as {meaning reconstruction from distributed evidence}, highlighting visual grounding and cultural-pragmatic adaptation as distinct model bottlenecks.

Abstract: LLM-based evaluators have rapidly become a standard tool for machine translation evaluation, yet the design choices that shape their behavior remain poorly understood. We present a systematic study of how instruction design, few-shot demonstrations, backbone model choice, and multi-step workflows affect span-level and ranking-based MT evaluation. We collect new expert RATE annotations on WMT24 and WMT25 English-Russian translations, with transfer validation on MQM- and ESA-annotated benchmarks across five additional language pairs. Our experiments reveal four robust findings: instruction structure shapes evaluator behavior, high-quality demonstrations matter more than quantity, frontier backbones exhibit distinct evaluator profiles (recall-, precision-, or ranking-oriented), and fusing outputs across diverse backbones outperforms single-evaluator multi-call setups. Although evaluator design exposes a large combinatorial space, several discovered design principles transfer consistently across annotation protocols and language pairs. We synthesize these findings into ORBIT, a family of LLM-based MT evaluators. Our strongest configurations achieve expert-level performance in error span detection and in Accuracy and Fluency ranking at roughly 20x lower cost than expert human annotation.

Abstract: Quality-Aware Decoding (QAD) improves machine translation through test-time scaling, generating multiple candidate translations and selecting the final output with a quality-aware decoder. Usually the same candidate budget N is spent on every source sentence, assuming that all inputs require the same amount of search. We show that this assumption does not hold and that the quality gain obtained from additional candidates varies across sentences. We study search utility through three quantities: total scaling gain, how much quality a sentence can gain from a much larger candidate pool; marginal scaling gain, the expected gain from generating one additional batch of candidates; and the sampling saturation point, the candidate budget required to realize most of the total scaling gain. We investigate two sources of information for predicting these quantities: signals available before candidate-set expansion, and statistics computed from the candidate pool accumulated during search. We find that sentence-level signals correlate with total scaling gain but are less informative about the sampling saturation point. In contrast, properties of the current candidate pool provide useful signals of marginal scaling gain. Finally, we show that these signals can identify sentences that obtain larger gains when additional search is selectively applied.

Abstract: Language models (LMs) often exhibit successful cross-lingual transfer through continued (pre)training, still their underlying process remains unclear. One possibility is that LMs can leverage the evidence toward language-universal principles, namely, hierarchical structure, from a source language (L1) to a new language (L2), while many other non-linguistic explanations, e.g., improved initialization of position embeddings, can be potentially possible. In this study, to quantify the cross-linguistic transferability of linguistic evidence in cross-lingual LM training, we introduce a new experimental design, namely, a cross-lingual poverty-of-stimulus (CPOS) setting that ablates how LMs' hierarchical generalization in L2 can be led by structural evidence provided in L1. We contrast counterfactual L1 and L2 learning scenarios where the L2 corpus is ambiguous between linear and hierarchical grammar rules, while the amount of positive evidence for the hierarchical rule is controlled on the L1 side. Our experimental results show that LMs can partially, yet not perfectly, transfer L1 structural evidence to L2, yielding only a small but detectable hierarchical bias, while the strength of transfer depends on language pairs. We further ablate the conditions for successful transfer, motivated by recent reports on the substantial impact of vocabulary overlap. We indeed find that vocabulary overlap is crucial even for the cross-lingual transfer of hierarchical syntactic generalization. This suggests that this also explains the language-pair-dependent tendencies in the main experiments and suggests that the transfer of linguistic evidence in continued new-language learning is only conditionally achievable.

Abstract: Pixel language models process text as rendered images instead of a finite vocabulary, which gives them strong cross-lingual transfer abilities. However, not much is known about how these abilities are influenced by larger-scale pretraining. In this paper, we train a family of MPIXEL models along three axes of variation: more languages, more scripts, and larger model capacity under fixed per-language compute. In evaluations on five tasks covering 83 languages, we find that scaling acts locally: each axis improves the added languages and scripts, while increased capacity mitigates the curse of multilinguality from pretraining on more languages. Transfer to unseen scripts improves with multilingual pretraining, but does not further improve with scaling. On typologically distant low-resource languages, our models outperform subword and character-level baselines across many writing systems. We also show that larger pixel models reserve more layers for visual processing, and that cross-lingual alignment weakens as more languages are added. Overall, our findings characterize how multilingual scaling improves pixel language models.

Abstract: Semantic map models (SMMs) construct a graph that reflects conceptual similarity by comparing cross-lingual form-meaning mappings. Experts build such models not only by obeying the data constraint, i.e., the connectivity hypothesis, which states that conceptual nodes belonging to the same form must be connected, but also by incorporating their prior semantic knowledge. Current automatic methods ignore this knowledge, which harms both accuracy and efficiency by enlarging the search space of candidate graphs. In this paper, we fuse the distributional semantic knowledge hidden in the internal representations of language models into a top-down, data-driven algorithm. This expert-like behavior re-sorts the candidates in a huge search space, guided by conceptual semantics similarity. Experiments on a self-collected benchmark of 10 datasets and comprehensive metrics show that our method builds SMMs both effectively and efficiently.

Abstract: Adapting a pretrained machine translation (MT) model to new terminology usually requires new parallel data and another round of fine-tuning, which is costly for low-resource languages. We ask whether controllability can be added to MT, and the fine-tuning burden reduced, without degrading the pretrained model. GLOSS-MT looks up the source's noun phrases in an offline English-Twi dictionary, replaces every phrase found by its Twi translation in place, and fine-tunes a translation model once to translate these mixed inputs. We apply the same data and pipeline to NLLB-200 (600M, 1.3B, 3.3B), OPUS-MT en-tw and TranslateGemma 4B. On a document-level English-to-Asante Twi benchmark (84 documents, four domains), every fine-tune scores above its pretrained base in chrF2 (+2.85 to +5.48), and the three NLLB fine-tunes are the best open systems (50.12-50.80). A direct controllability test on human-translated Bible passages shows that the three fine-tunes tested write 79-97% of the dictionary terms, and that replacing every dictionary entry by a different term changes the output accordingly (70-97% of the new terms appear) with no retraining. We release the models, data and code.

Abstract: While deriving text embeddings directly from frozen LLMs has emerged as a promising direction, existing methods focus almost exclusively on English, leaving multilingual settings largely unexplored. In this work, we present the first systematic study of multilingual training-free text embeddings in LLMs. Our analysis reveals a language-conditioned embedding geometry in which semantically equivalent sentences remain misaligned across languages. We hypothesize that language-specific variation can be disentangled from semantic content to improve embedding quality. To this end, we propose Paired-Residual Subspace Removal (PRSR), a training-free inference-time intervention that steers LLM representations toward a more language-invariant semantic space. PRSR first constructs paired residuals from multilingual parallel sentences to capture language-specific deviations. It then estimates layer-wise low-rank subspaces from these residuals and suppresses the corresponding components in the last-token hidden states. We conduct more than 200 monolingual evaluations across 18 languages using MTEB and MMTEB, along with cross-lingual bitext mining between English and over 200 other languages. Empirical results demonstrate that PRSR consistently improves the performance of existing training-free baselines in both monolingual and cross-lingual settings across different LLMs.

Abstract: Recent multilingual large language models achieve strong translation performance, yet the internal mechanisms responsible for low-resource multilingual translation remain insufficiently understood. In this paper, we study translation circuits in multilingual decoder-only LLMs for Hindi-to-Urdu and Hindi-to-Sindhi (Devanagari). We propose TITAN, a mechanistic framework that combines causal intervention, sparse feature decomposition, and targeted preference optimization to identify and optimize crucial translation components in the transformer. Unlike conventional full-parameter fine-tuning, TITAN restricts optimization to sparse, causally important circuits, enabling scalable and parameter-efficient adaptation while updating fewer than 5% of model parameters. Experimental results show that TITAN achieves BLEU improvements of 33.75% on Hindi-to-Urdu and 45.86% on Hindi-to-Sindhi over the strongest multilingual baselines. Human expert evaluation further confirms substantial gains in adequacy and fluency across both language pairs. Additional analyses reveal that multilingual translation is implemented through sparse, structured circuits operating over a shared semantic subspace rather than through an English-centric pivot representation. Overall, our findings demonstrate that mechanistic circuit discovery can directly improve low-resource multilingual translation while providing interpretable insights into multilingual transformer behavior.

Abstract: We present PatTransBench, a 12,000-item Chinese-English/Japanese/Korean patent machine translation (PatMT) benchmark with a four-dimensional evaluation protocol. The benchmark comprises 6,000 short-text (sentence/claim) and 6,000 long-text (paragraph) pairs, evenly split across three language pairs and both translation directions, built via IPC-section-stratified sampling and dualthreshold quality gating from upstream patent-family parallel corpora. The evaluation protocol spans literal n-gram overlap (BLEU, ROUGE-L, chrF), cross-lingual semantic textual similarity (CLSTS), 38-probe-language consistency testing, and serving performance under concurrency. A case study on PatMT2-7B and its fp8/AWQ4/AutoRound-4bit variants reveals direction-dependent quantization effects masked by aggregate scores and non-monotonic language-failure patterns across quantization schemes that are invisible to single-metric evaluation. We plan to release the benchmark, protocol, and case-study results upon publication to support reproducible PatMT evaluation.

Abstract: India's Eighth Schedule recognises 22 languages, yet parallel data is uneven across them, and many language pairs, especially among lower-resource languages such as Bodo, Dogri, Santali and Kashmiri, lack direct bitext. We present BharatSetu 22-Indic 1M, a row-aligned table of 984,764 English sentences, each paired with machine translations into all 22 scheduled languages (21,664,808 Indic cells), generated with the sarvamai/sarvam-translate model family. Because every Indic cell in a row translates the same English sentence, any two columns form a translation pair: 462 directed Indic-to-Indic pairs, or 506 including English, without direct bitext. We call these pivot-derived pairs and propose a confidence layer that lets users choose how strictly each pair is connected. The corpus is machine-generated and not human-verified; quality is expected to vary by language, domain and sentence style, and pivot-derived pairs can inherit errors from both sides. We therefore release BharatSetu as a machine layer and propose a human-correction layer keyed to the same row identifiers, so that each corrected column upgrades every pair that uses it, together with an evaluation protocol. We report no translation-quality results.

Abstract: Feedback presentation shapes how LLMs use evaluator judgments during translation refinement. We introduce a machine translation protocol that keeps the evaluator score, main issue, and recommendation fixed while varying feedback form and supporting detail. Across seven models and six translation directions, low-load feedback outperforms high-load feedback in all 42 translator-direction settings under xCOMET. On the 29,400-record rewriting subset, mean xCOMET scores multiplied by 100 are 83.30 for low-load feedback and 80.67 for high-load feedback; the full-matrix original-translation reference mean is 83.46. MetricX-24 supports the low-load advantage on a stratified subset. A no-feedback second pass yields a small improvement over the original translations, and direct evaluator prompting reproduces the low-load advantage without a separate rewriting stage. These results show that compact feedback limits the quality loss from evaluator-guided refinement in the evaluated high-resource MT settings.

Abstract: Diffusion language models generate tokens in parallel, but still require multiple sampling steps to produce high-quality text. In this paper, we investigate whether a model can learn to generate an entire sequence in a single forward pass. We propose Seq-Drifting, a framework that applies drifting to sequences of continuous token representations and directly learns a map from Gaussian noise to text. Unlike existing DLMs that derive the supervision at each position directly from the observed token in a training sequence, Seq-Drifting builds a drift field whose position-wise targets adapt to the current generated context. At each position, attraction guides each output toward a set of contextually plausible token representations, while repulsion discourages collapse across samples and repetition within sequences. The generator learns from this drift field during training and produces all tokens jointly in one pass at inference. Empirically, Seq-Drifting achieves a generative perplexity of 36.01 on LM1B in one step and performs competitively in both unconditional and conditional generation. Experiments on translation, summarization, and reasoning further demonstrate the applicability of the framework beyond open-ended generation. These results suggest that drifting over token sequences offers a promising path toward effective one-step text generation. Our code is available at https://github.com/jyliu-ai/Seq-Drifting.

Abstract: Neural Machine Translation (NMT) models, while capable of producing highly fluent outputs, remain vulnerable to hallucinations, which are translations that are natural yet semantically unrelated to the source. This vulnerability is acute in low-resource settings like Sinhala-to-English, where weak cross-lingual alignment leads to hallucinations. This paper introduces a framework for reference-free hallucination detection in this language pair. We present a 45,000-sample synthetic dataset generated through a probabilistic chain of five linguistically motivated corruption strategies, with a semantic rescue mechanism that uses character-level similarity to distinguish hallucinations from morphological variants. We fine-tune mDeBERTa-v3 for token-level sequence labelling, reaching a token-level F1 of 0.841 +/- 0.001 over three seeds on a source-disjoint test set, and study a three-signal ensemble integrating neural risk scores, sequence log-probabilities, and cross-lingual semantic embeddings (LaBSE). A source-ablation control shows that the detector relies on the Sinhala source rather than on surface artefacts of the corruption process: shuffling or removing the source reduces sentence-level AUROC from 0.970 to chance. We benchmark eight NMT systems spanning five model families and find that detector firings vary by an order of magnitude across architectures.

Abstract: Large Language Models (LLMs) are considered to be a more efficient and cost-effective alternative to human judgment for Machine Translation (MT) evaluation. With MT evaluation spanning a large number of language pairs, domains and levels of annotation granularity, LLMs must be thoroughly evaluated across these dimensions before being reliably used as alternatives to human evaluation. In this paper, we evaluate the performance of LLMs for two prominent MT quality evaluation schemes: Multidimensional Quality Metrics (MQM) and Error Span Annotation (ESA) by comparing their agreement with human annotators. We present results on a long-context test set of 70 language pairs and the publicly available WMT23 and WMT25 data, investigating both score and error span annotation agreement across a variety of language pairs and domains. Our results show that while LLM agreement with human annotators exceeds agreement between human annotators for some evaluation tasks, both vary substantially across annotation schemes, language pairs and domains and remain unreliable for most settings. Furthermore, we identify challenges facing both human and LLM annotators: humans are particularly challenged by fine-grained MQM annotations and low-resource language pairs, while LLMs struggle with minor errors, wrong language variants and error span annotation. Our results highlight the potential for improvement for both human and LLM annotation performance, possibly through human-LLM collaborative annotation pipelines that address the reliability issues identified in this work.

Abstract: Representations of translated sentences are similar in the inner layers of multilingual language models -- an observation connected to the platonic representation hypothesis, yet unexplained theoretically. We provide an explanation based on the assumption that data have a hidden hierarchical structure whose abstract levels are shared across languages while surface levels are modality- or language-specific. Concretely, we generate synthetic languages from probabilistic context-free grammars sharing upper-level but not lower-level production rules. In this setting the Bayes-optimal next-token predictor is belief propagation (BP); encoding its messages in successive layers yields analytical predictions that agree well with transformers trained on the same data. The framework explains why cross-lingual similarity peaks in middle layers, coexists with language-specific structure, and strengthens with language proximity, model quality and data exposure. It distinguishes similarity (shared neighborhood geometry) from alignment (shared coordinates), showing that the latter occurs when code-switched data, i.e. mixed-language sentences, are abundant enough. It further predicts that subtracting from each layer the component linearly predictable from the preceding one increases cross-lingual similarity, which we confirm in pretrained LLMs.

Abstract: Dialectal machine translation remains challenging due to limited data and strong linguistic variation not captured by standard benchmarks, which often assume standardized and well-edited text. We study Polish-Silesian MT using neural and rule-based systems, evaluating on SiLTT - a new Pol-Szl testset, alongside established BOUQuET and FLORES benchmarks. Results show our rule-based system is consistently strongest on SiLTT and BOUQuET datasets and that TranslateGemma fine-tuned on a curated dataset improves over strong neural baselines but does not surpass the rule-based system in dialectal settings. We release SiLTT and our best neural model to support further research.

Abstract: Cross-lingual transfer describes how knowledge in a source language benefits a target language. Measuring it quantitatively requires broad multilingual pre-training, as prior work has done with cross-lingual transfer matrices. We ask whether transfer is predictable from freely available typological features, and whether the prominence of high-resource source languages reflects typology or data quality and quantity. We show that typological databases contain cheap and dense signals about cross-lingual transfer. Our typology-only random forest on a 24-language prior-work transfer matrix scores leave-one-language-out $ρ{=}0.705$ and $R^2{=}0.49$, beating a non-typological control at $ρ{=}0.62$, which verifies the ability of typology-only predictions to reconstruct costly measured cross-lingual transfer. The signal survives leave-one-script-out and leave-one-family-out protocols, so script and family confounding do not explain the effect. By decomposing the transfer into a typology term and a resource-and-script bias term, we find the best-source ranking sensitive to this bias. In contrast, typology is not affected by this bias, which makes it a zero-compute screening tool that replaces hundreds of training runs with a model fit. Our code is available \href{https://github.com/dharmsen/typo-x-ling-transfer}{here}.

Abstract: Pre-trained language models (PLMs) with encoder-based architectures have shown impressive capabilities in zero-shot cross-lingual transfer for various language understanding tasks. However, applying this technique to dependency parsing remains a significant challenge due to its syntactic nature. To boost model generalizability across linguistic typologies, we propose a cross-lingual unsupervised bootstrapping method to improve syntactic knowledge within the PLM. We show that our method achieves a significant improvement in zero-shot parsing performance in low-resource languages. Analysis of these bootstrapped models uncovers increased robustness in recognizing syntactic structures, evidenced by higher scores in parameter-free tree probing tests.

Abstract: Among closely related North Germanic languages, the quantification of language distance has traditionally relied on qualitative methods, lacking a unified multi-dimensional computational framework. Multilingual pretrained models based on the Transformer architecture can map texts from different languages into a shared vector space, enabling quantitative measurement of language distance. This paper focuses on the three North Germanic languages---Danish, Norwegian (Bokmål), and Swedish---and proposes a three-metric quantitative framework based on pretrained language models: (1)~sentence-level semantic distance, computed as cosine similarity between LaBSE and mBERT encodings of parallel sentences; (2)~orthographic fragmentation rate, measuring subword tokenization efficiency when cross-applying monolingual BERT vocabularies to parallel texts; (3)~MLM predictability, comparing prediction confidence and entropy in masked language modeling using mBERT across languages. Using 150 trilingual parallel sentence triplets from the Tatoeba corpus as controlled samples, we obtain consistent distance rankings on two independent models: LaBSE: da--no $0.012 < $ no--sv $0.016 < $ da--sv $0.020$; mBERT: da--no $0.016 < $ no--sv $0.045 \approx $ da--sv $0.046$. This ranking is consistent with the historical linguistic conclusion that ``400 years of Danish rule over Norway (1380--1814) led to highly cognate written languages.'' The three metrics---semantic, orthographic, and predictability---converge on the same conclusion, providing a reproducible computational framework for the quantitative study of distance among closely related languages, extensible in principle to more branches of the Indo-European language family, pending validation on additional language groups.

Abstract: Classical texts aligned with their translations support machine translation, retrieval and computational research, but evidence comparing alignment workflows is scattered. This study compares seven systems on 452 texts in Pali, Sanskrit, Mishnaic Hebrew and Tibetan, comprising 9,833 human-aligned units: four embedding pipelines, a direct LLM call, an autonomous agent, and the agent revised by an independent auditor. Generative workflows recover 93-94% of reference correspondences, against at most 77% for embeddings. A ceiling analysis shows that sentence boundaries make some references unrepresentable by the embedding pipelines. Reference recovery is similar across generative workflows: the agent's advantage is 0.5 percentage points (95% CI -0.02 to 1.17), and auditing adds no established benefit. Agents nevertheless produce structurally valid output for all 452 texts, against 437 for direct calls. A blinded three-LLM panel assesses every generative mismatch against the source and human reference. Most mismatches are labelled defensible editorial variation; consensus major-error labels cover only 0.06-0.14% of units. The panel labels significantly fewer residual defects for agents than direct calls (0.7% versus 1.4%), suggesting that reference recovery alone understates alignment quality. On ten long Pali discourses taken as published online, agents and audited agents raise recovery from the direct call's 71% to 84% and 92%. Identical reference-located chunks bring all three to 93%. Agents thus improve structural reliability and reduce judged defects on short passages, while their large recovery advantage on long documents disappears after chunking. In this setting, independent auditing offers little measurable additional benefit on prepared passages.

Abstract: Machine translation (MT) for Indian languages remains constrained by the limited availability of high-quality, Indic-centric parallel corpora and evaluation benchmarks. Existing multilingual resources are largely constructed from English-pivot content and often fail to capture the linguistic diversity, cultural complexity, and domain-specific characteristics of Indian languages. We present COILD, an Indic-centric parallel corpus comprising over 1.16 million human-translated and human-verified sentence pairs, covering 20 Indian language pairs across the Indo-Aryan, Dravidian, Tibeto-Burman, and Austro-Asiatic language families. The corpus is built entirely from original Indian language sources collected from licensed repositories spanning eight domains with direct real-world applicability. Furthermore, we introduce a domain-centric benchmark comprising 2,000 expert-verified sentences to enable consistent multilingual and cross-lingual evaluation across Indian language pairs. To validate the effectiveness of COILD, we fine-tune two representative multilingual neural machine translation models, IndicTrans2-Distilled and NLLB-200. Experimental results demonstrate consistent improvements across language pairs, domains, automatic evaluation metrics, and human evaluation, highlighting the effectiveness of high-quality Indic-centric supervision. COILD provides a valuable training and evaluation resource for advancing multilingual machine translation and future multilingual language models for Indian languages.

Abstract: Fine-tuning large language models on parallel data improves translation quality but can cause catastrophic forgetting. Mitigation methods are generally evaluated by retention on general benchmarks. We ask whether these findings transfer to machine translation (MT) fine-tuning and to MT-specific instruction following (MT-IF): instructions that modify a translation, such as formality, grammatical gender, and length control. We compare methods anchored to auxiliary data, to model outputs, and to the base model parameters, first in a screening study with Llama 3.2 1B Instruct, then on Llama 3.1 8B Instruct fine-tuned on bidirectional Arabic-English or Spanish-English data. Elastic Weight Consolidation preserves general capabilities best in both stages; on the 8B Spanish model the average score on general benchmarks drops 1.7 points versus 11.0 for standard fine-tuning, yet its scores for formality and grammatical gender control remain close to standard fine-tuning. Only data mixing with control-task examples preserves these controls, but its gains do not transfer to unseen prompts for the same task.

Abstract: Cross-lingual legal question answering must retrieve statutes across languages while preventing unsupported legal claims. We introduce a bilingual evaluation suite of 231 Vietnamese--English question--answer pairs from Vietnamese labour law. Of these, 75 are additionally annotated for five challenging legal reasoning phenomena. We evaluate a verifier-guided pipeline that decomposes answers into claims, checks citation reachability and entailment, and corrects citation failures and contradictions. We also introduce six automatic diagnostics for faithfulness to retrieved evidence, covering citations, modality, exceptions, procedures, conclusions, and evidential support. Experiments show that learned-sparse retrieval performs poorly for English-to-Vietnamese retrieval (R@5~=~0.032), whereas dense retrieval reaches 0.358 and slightly outperforms hybrid retrieval. Translation placement has no statistically detectable effect on these automatic diagnostics in our controlled comparison and supporting sensitivity analyses. Verifier-guided correction improves citation preservation by $0.022$--$0.034$ at the system level but produces no reliable gains in the remaining dimensions. Human evaluation further shows that the automatic diagnostics do not fully align with human judgements of answer quality.

Abstract: 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.

Abstract: Large-scale machine-translation (MT) systems are typically evaluated on random samples from a corpus whose distributional composition is an artifact of how it was assembled. Such a sample inherits the phenomena the collection happens to contain rather than the full space a system must handle, spanning rule-governed conventions (terminology, punctuation, currency formatting) and context-dependent phenomena (tone, honorifics, document-level coherence), and thus provides no coverage guarantee for assessing robustness. We propose TACTICS (Taxonomy-Aware Coverage-opTimized Intelligent Corpus Sampling), which recasts coverage as an explicit objective. TACTICS induces a hierarchical taxonomy from a locale style guide, classifies segments against it, and selects a fixed-budget subset jointly optimizing coverage of rare categories, document-level coherence, and distributional fidelity to the full corpus. Applied to MT evaluation across four translation directions, TACTICS improves coverage of rare categories over lexical and embedding-based selection. By targeting the phenomena that separate systems, TACTICS makes a fixed evaluation budget go further, recovering the true system ranking from far fewer segments than random sampling wherever a real quality gap exists and never signaling a difference where none exists.

Abstract: This paper presents a Belarusian-specific data-cleaning pipeline and fine-tuning for English-Belarusian machine translation. Our cleaning pipeline distinguishes itself from others by employing a correction tool that addresses the issue of the two orthographies of the Belarusian language, noise in the training data, interference from other languages and other misspelling issues common in Belarusian on the internet. A matched ablation on unfiltered training data shows substantial benefits from filtering for all fine-tuned models, with the LLM-based models gaining roughly twice as much from filtering as the dedicated encoder-decoder MT system, supporting the view that for Belarusian MT one of the primary bottlenecks is data quality.

Abstract: There is a tradeoff in machine translation meta-evaluation between prioritizing alignment with adequacy versus fluency. The balance depends on the combination of translation systems in the meta-evaluation dataset. This system set is a small, filtered sample whose characteristics change heavily across years and language pairs; it does not represent the true system distribution. Consequently, the adequacy-fluency balance is often unrepresentative and subject to change. For sensitive domains, controlling this balance is critical. We expose this balance as a tunable choice. To achieve a target balance, we reweight existing systems while minimizing distortion from uniform weighting, ensuring the evaluated systems remain real and representative. We provide an exact optimization algorithm with theoretical guarantees and pruning mechanisms to compute these weights. To validate meta-evaluation internal consistency, we design a scorer-augmentation framework that establishes a known relative identity for the scorers. Results demonstrate that our reweighting method effectively controls the adequacy-fluency balance and preserves the internal consistency, outperforming prior approaches. Finally, we analyze the performance of popular scorers across a sweep of this parameter.

Abstract: Human corrections identify editable spans, but the examples receiving corrections may come from a selective feedback channel. We analyze this interaction at a fixed model checkpoint by decomposing a localized gradient into edited and retained untouched components. Squared relative selection bias is a ratio of quadratics whose derivative has the sign of an explicit quadratic polynomial. Localization can increase, decrease, or nonmonotonically change this diagnostic; its direction depends on component biases and geometry. Oracle importance weighting recovers the population mean for each fixed localization objective, but these objectives have different targets. Against one common full-gradient target, we derive the finite-sample mean-squared error, an analytic optimal retention coefficient, and a fixed-clipping extension. Exact finite-population calculations and 10,000 Monte Carlo repetitions per sample size verify the identities and counterexamples. Public human-post-edit experiments use two translation directions and pretrained models, with declared synthetic selection. An English-German extension differentiates 73.89 million native parameters. In all three declared settings, hard localization has higher relative bias but lower absolute bias than full retention. Untouched-component biases are nonzero and selected component means have negative inner products, so the general criterion applies where the simple unbiased/aligned explanation fails. Output-bias diagnostics show the same endpoint ordering of relative bias across both directions, with one interior maximum. Mechanisms reuse each language's records and include a mixture; they are not independent replications. The evidence separates relative amplification from absolute gradient error and establishes estimation properties, without inferring translation-quality gains or identifying actual complaint propensities.

Abstract: Sanskrit fuses case, number, person and tense into word endings and chains clauses into compounds, so it is information-dense per word. Whether that density survives subword tokenization is a separate question, to be asked per unit of meaning rather than per word. On identical FLORES-200 devtest content, Sanskrit costs 1.774-2.187 times the English tokens under deployed tokenizers with vocabularies of 200,019 ids or more, but only 1.325-1.353 times the Hindi tokens. Against a deployed English tokenizer, Sanskrit-trained BPE arms then look cheaper per proposition than English on contemporary prose (0.887). Against a matched English control, the same algorithm and vocabulary trained on the English side of the same corpus, that flip disappears: at 32,000 and 64,000 pieces all 8 matched pairs, each size-matched arm against both a pair-matched and a byte-matched control, sit above 1.0 on prose with 95% intervals excluding it. The gap closes as the vocabulary grows: at 128,000 pieces the BPE pair reads 0.983 in domain while staying above parity out of domain (1.025) and on FLORES (1.116). The ratio factorises into a character-length ratio and a tokens-per-character ratio, the second near 1 throughout: what survives matched tokenization is character-level length, which Sanskrit prose lacks over English in SLP1 (1.028) and Sanskrit verse has (0.596). The robust statement is about deployed practice: on contemporary prose and on FLORES, with the Sanskrit side in SLP1 against the deployed o200k English pivot, Sanskrit costs 1.831-2.899 English tokens per proposition under the tokenizers people actually ship. Code, the results snapshot and every table here are public.

Abstract: Advanced large language models (LLMs) with long context windows can substantially reduce input truncation in document-level machine translation (DocMT). However, direct Doc2Doc translation remains prone to n-gram repetition and progressive quality degradation. A common remedy is to segment the document into finer-grained chunks. Nonetheless, conventional rule-based chunking approaches fail to handle the length distribution mismatch between training and inference. To address this, we introduce Fixed-Range Chunking (FRC), utilizing dynamic programming to partition documents into chunks within a predefined length interval. By consistently applying FRC during training and inference, the input documents of any length are mapped to the same length distribution, substantially reducing train-test length mismatch. Centered on FRC, we propose a lightweight dual-boundary matching algorithm for chunk alignment, alongside four distinct training strategies. Experimental results show that FRC-based fine-tuning substantially improves 7B LLMs over direct Doc2Doc fine-tuning and outperforms existing DocMT methods on IWSLT2017. We further construct GlobVDoc, a 10-language test set independent of mainstream DocMT training sources, and show that FRC improves out-of-distribution document translation.

Abstract: We describe six submissions under the team name ESTS to the unconstrained WMT26 Model Compression Shared Task for English--Simplified Chinese and English--Egyptian Arabic. We submit three compression operating points per translation direction, all derived from GPT-OSS-20B. We use task-specific routing mass to rank experts and cross-lingual routing divergence to allocate retained capacity across layers, then physically remove low-importance experts. The resulting specialists are recovery-tuned on GPT-5.1-generated synthetic translation data and further compressed by applying MXFP4 quantization to the retained expert projection weights. We additionally implement a robust inference system for the instruction-conditioned WMT26 setting, including category inference, output validation, retries, segmented fallback, and source-owned JSON reconstruction. Across our six submissions, parameter counts range from 4.186B to 7.770B and packed artifact sizes from 4.55 to 6.33~GiB. Internal xCOMET-XL evaluation using GPT-5.1 pseudo-references provides an internal comparison across the submitted compression operating points.

Abstract: Multilingual large language models (mLLMs) contain a small fraction of feed-forward neurons that are sensitive to particular languages, commonly termed language-specific neurons. Existing work measures language specificity using the entropy of each neuron's language-wise probabilities of being active, where a neuron is considered active when its activation value is positive. However, this approach may not fully capture the multilingual nature of mLLMs, where language representations are distributional and mutually related. We propose Distribution-aware Language Neuron selection, which leverages pairwise relationships between per-language activation distributions over the full activation range, including negative values. Specifically, we quantify each neuron's language specificity by clustering languages using pairwise overlap coefficients between their activation distributions. Across two mLLMs and two held-out corpora, our identifier more effectively isolates language-specific causal effects, yielding up to 4.9$\times$ higher on-target language damage per neuron while preserving off-target language performance.

Abstract: Terminology evaluation in machine translation (MT) usually assumes a single correct target form per source term. However, human translators routinely introduce variation that current metrics penalize as inconsistency. We examine how to account for this variation in document-level MT evaluation of English-French scientific translation, combining glossary-based accuracy, translation consistency, and a new cross-term variation (CTV) diagnostic measure that tests whether variation relationships are preserved across languages. Based on analyses of two parallel corpora, translated by four MT systems, we find that (1) MT systems generate less target-side variation than human translators; (2) transfer patterns strongly depend on the variation type; (3) consistency rankings vary with the choice of metric; and (4) constraining MT with a glossary improves accuracy and consistency but degrades CTV by suppressing valid variation. We argue for variation-aware evaluation that conditions consistency penalties on whether target-side variation mirrors source-side variation.

Abstract: Large language models (LLMs) are commonly associated with the distributional hypothesis, according to which (1) semantic meaning is grounded in distributional patterns of linguistic context, and (2) knowledge of cross-linguistic distributional correspondences allows for successful translation. This paper rejects the first claim as a causal inversion: linguistic distributions reflect patterns arising from meaning-making practices rather than constituting their source. At the same time, it accepts the second claim, arguing that translation -human or machine - can succeed without requiring access to meaning or reference. Knowledge of interlingual distributional correspondence and their inferential organization may be sufficient for translation. The paper develops an ecological-enactivist perspective, according to which reference and meaning are grounded in agent-environment interaction and stabilized through action-grounded concepts, forms of world-involving cognition that current LLMs do not possess.