Sanchit Ahuja

Sanchit Ahuja, Terra Blevins · uw

While continual pretraining~(CPT) is a practical way to extend large language models to new languages, naïve finetuning on targeted data erodes existing capabilities through catastrophic forgetting. Organizing training around language families reduces cross-language interference but cannot alone prevent forgetting of the general knowledge needed for downstream tasks. We link this forgetting to parameter drift in multilingual CPT and present a suite of five layer-aware parameter alignment strategies: hard layer freezing, soft regularization, post-hoc weight reversion, and model merging. We systematically compare our alignment strategies against two unregularized CPT baselines on benchmarks spanning 32 training languages from five language families, plus held-out languages, across four evaluation axes: perplexity, reading comprehension, physical reasoning, and translation. Parameter alignment substantially reduces forgetting at minimal cost to language acquisition: layer freezing and regularization best preserve comprehension, whereas post-hoc reversion yields the strongest translation gains. Together, these results map the acquisition--forgetting frontier for family-expert CPT and offer practical deployment guidelines pairing each strategy to the tasks it best serves.

Sanchit Ahuja, Divyanshu Aggarwal, Varun Gumma et al. · cmu, deepmind

There has been a surge in LLM evaluation research to understand LLM capabilities and limitations. However, much of this research has been confined to English, leaving LLM building and evaluation for non-English languages relatively unexplored. Several new LLMs have been introduced recently, necessitating their evaluation on non-English languages. This study aims to perform a thorough evaluation of the non-English capabilities of SoTA LLMs (GPT-3.5-Turbo, GPT-4, PaLM2, Gemini-Pro, Mistral, Llama2, and Gemma) by comparing them on the same set of multilingual datasets. Our benchmark comprises 22 datasets covering 83 languages, including low-resource African languages. We also include two multimodal datasets in the benchmark and compare the performance of LLaVA models, GPT-4-Vision and Gemini-Pro-Vision. Our experiments show that larger models such as GPT-4, Gemini-Pro and PaLM2 outperform smaller models on various tasks, notably on low-resource languages, with GPT-4 outperforming PaLM2 and Gemini-Pro on more datasets. We also perform a study on data contamination and find that several models are likely to be contaminated with multilingual evaluation benchmarks, necessitating approaches to detect and handle contamination while assessing the multilingual performance of LLMs.

Mubashara Akhtar, Anka Reuel, Prajna Soni et al. · meta-ai

Artificial Intelligence (AI) benchmarks play a central role in measuring progress in model development and guiding deployment decisions. However, many benchmarks quickly become saturated, meaning that they can no longer differentiate between the best-performing models, diminishing their long-term value. In this study, we analyze benchmark saturation across 60 Large Language Model (LLM) benchmarks selected from technical reports by major model developers. To identify factors driving saturation, we characterize benchmarks along 14 properties spanning task design, data construction, and evaluation format. We test five hypotheses examining how each property contributes to saturation rates. Our analysis reveals that nearly half of the benchmarks exhibit saturation, with rates increasing as benchmarks age. Notably, hiding test data (i.e., public vs. private) shows no protective effect, while expert-curated benchmarks resist saturation better than crowdsourced ones. Our findings highlight which design choices extend benchmark longevity and inform strategies for more durable evaluation.

Sanchit Ahuja, Kumar Tanmay, Hardik Hansrajbhai Chauhan et al.

Despite the remarkable success of large language models (LLMs) in English, a significant performance gap remains in non-English languages. To address this, we introduce a novel approach for strategically constructing a multilingual synthetic instruction tuning dataset, sPhinX. Unlike prior methods that directly translate fixed instruction-response pairs, sPhinX enhances diversity by selectively augmenting English instruction-response pairs with multilingual translations. Additionally, we propose LANGIT, a novel N-shot guided fine-tuning strategy, which further enhances model performance by incorporating contextually relevant examples in each training sample. Our ablation study shows that our approach enhances the multilingual capabilities of Mistral-7B and Phi-3-Small improving performance by an average of 39.8% and 11.2%, respectively, across multilingual benchmarks in reasoning, question answering, reading comprehension, and machine translation. Moreover, sPhinX maintains strong performance on English LLM benchmarks while exhibiting minimal to no catastrophic forgetting, even when trained on 51 languages.

Sanchit Ahuja, Praneetha Vaddamanu, Barun Patra

Despite recent advances in Language Reasoning Models (LRMs), most research focuses solely on English, even though many models are pretrained on multilingual data. In this work, we investigate: Is English the most token-efficient language for reasoning? We evaluate three open-source RLMs: DeepSeek R1, Qwen 2.5 and Qwen 3, across four math datasets and seven typologically diverse languages. We find that reasoning in non-English languages not only reduces token usage, but also preserves accuracy. These gains persist even after translating the reasoning traces into English, suggesting genuine shifts in reasoning behavior rather than surface-level linguistic effects. The extent of improvement, however, depends on the models multilingual strength. Our findings motivate a broader view of reasoning in language models, highlighting the potential of multilingual reasoning and the importance of strong multilingual foundations. The code for our work can be found: https://github.com/microsoft/EfficientXLang.

Pranjal A. Chitale, Varun Gumma, Sanchit Ahuja et al. · microsoft-research

Developing AI systems that operate effectively across languages while remaining culturally grounded is a long-standing challenge, particularly in low-resource settings. Synthetic data provides a promising avenue, yet its effectiveness in multilingual and multicultural contexts remains underexplored. We investigate the creation and impact of synthetic, culturally contextualized datasets for Indian languages through a bottom-up generation strategy that prompts large open-source LLMs (>= 235B parameters) to ground data generation in language-specific Wikipedia content. This approach complements the dominant top-down paradigm of translating synthetic datasets from high-resource languages such as English. We introduce Updesh, a high-quality large-scale synthetic instruction-following dataset comprising 9.5M data points across 13 Indian languages, encompassing diverse reasoning and generative tasks with an emphasis on long-context, multi-turn capabilities, and alignment with Indian cultural contexts. A comprehensive evaluation incorporating both automated metrics and human annotation across 10k assessments indicates that generated data is high quality; though, human evaluation highlights areas for further improvement. Additionally, we perform downstream evaluations by fine-tuning models on our dataset and assessing the performance across 15 diverse multilingual datasets. Models trained on Updesh consistently achieve significant gains on generative tasks and remain competitive on multiple-choice style NLU tasks. Notably, relative improvements are most pronounced in low and medium-resource languages, narrowing their gap with high-resource languages. These findings provide empirical evidence that effective multilingual AI requires multi-faceted data curation and generation strategies that incorporate context-aware, culturally grounded methodologies.

Nedjma Ousidhoum, Shamsuddeen Hassan Muhammad, Mohamed Abdalla et al.

Exploring and quantifying semantic relatedness is central to representing language and holds significant implications across various NLP tasks. While earlier NLP research primarily focused on semantic similarity, often within the English language context, we instead investigate the broader phenomenon of semantic relatedness. In this paper, we present \textit{SemRel}, a new semantic relatedness dataset collection annotated by native speakers across 13 languages: \textit{Afrikaans, Algerian Arabic, Amharic, English, Hausa, Hindi, Indonesian, Kinyarwanda, Marathi, Moroccan Arabic, Modern Standard Arabic, Spanish,} and \textit{Telugu}. These languages originate from five distinct language families and are predominantly spoken in Africa and Asia -- regions characterised by a relatively limited availability of NLP resources. Each instance in the SemRel datasets is a sentence pair associated with a score that represents the degree of semantic textual relatedness between the two sentences. The scores are obtained using a comparative annotation framework. We describe the data collection and annotation processes, challenges when building the datasets, baseline experiments, and their impact and utility in NLP.

Agrima Seth, Sanchit Ahuja, Kalika Bali et al.

Generative models are increasingly being used in various applications, such as text generation, commonsense reasoning, and question-answering. To be effective globally, these models must be aware of and account for local socio-cultural contexts, making it necessary to have benchmarks to evaluate the models for their cultural familiarity. Since the training data for LLMs is web-based and the Web is limited in its representation of information, it does not capture knowledge present within communities that are not on the Web. Thus, these models exacerbate the inequities, semantic misalignment, and stereotypes from the Web. There has been a growing call for community-centered participatory research methods in NLP. In this work, we respond to this call by using participatory research methods to introduce $\textit{DOSA}$, the first community-generated $\textbf{D}$ataset $\textbf{o}$f 615 $\textbf{S}$ocial $\textbf{A}$rtifacts, by engaging with 260 participants from 19 different Indian geographic subcultures. We use a gamified framework that relies on collective sensemaking to collect the names and descriptions of these artifacts such that the descriptions semantically align with the shared sensibilities of the individuals from those cultures. Next, we benchmark four popular LLMs and find that they show significant variation across regional sub-cultures in their ability to infer the artifacts.

Nedjma Ousidhoum, Shamsuddeen Hassan Muhammad, Mohamed Abdalla et al.

We present the first shared task on Semantic Textual Relatedness (STR). While earlier shared tasks primarily focused on semantic similarity, we instead investigate the broader phenomenon of semantic relatedness across 14 languages: Afrikaans, Algerian Arabic, Amharic, English, Hausa, Hindi, Indonesian, Kinyarwanda, Marathi, Moroccan Arabic, Modern Standard Arabic, Punjabi, Spanish, and Telugu. These languages originate from five distinct language families and are predominantly spoken in Africa and Asia -- regions characterised by the relatively limited availability of NLP resources. Each instance in the datasets is a sentence pair associated with a score that represents the degree of semantic textual relatedness between the two sentences. Participating systems were asked to rank sentence pairs by their closeness in meaning (i.e., their degree of semantic relatedness) in the 14 languages in three main tracks: (a) supervised, (b) unsupervised, and (c) crosslingual. The task attracted 163 participants. We received 70 submissions in total (across all tasks) from 51 different teams, and 38 system description papers. We report on the best-performing systems as well as the most common and the most effective approaches for the three different tracks.

Yifei He, Alon Benhaim, Barun Patra et al.

We propose a novel scaling law for general-purpose decoder-only language models (LMs) trained on multilingual data, tackling the problem of balancing languages during multilingual pretraining. A primary challenge in studying multilingual scaling is the difficulty of analyzing individual language performance due to cross-lingual transfer. To address this, we shift the focus from individual languages to language families. We introduce and validate a hypothesis that the test cross-entropy loss for each language family is determined solely by its own sampling ratio, independent of other languages in the mixture. This insight simplifies the complexity of multilingual scaling and make the analysis scalable to an arbitrary number of languages. Building on this hypothesis, we derive a power-law relationship that links performance with dataset size, model size and sampling ratios. This relationship enables us to predict performance across various combinations of the above three quantities, and derive the optimal sampling ratios at different model scales. To demonstrate the effectiveness and accuracy of our proposed scaling law, we perform a large-scale empirical study, training more than 100 models on 23 languages spanning 5 language families. Our experiments show that the optimal sampling ratios derived from small models (85M parameters) generalize effectively to models that are several orders of magnitude larger (1.2B parameters), offering a resource-efficient approach for multilingual LM training at scale.

Sanchit Ahuja, Varun Gumma, Sunayana Sitaram · microsoft-research

Benchmark contamination refers to the presence of test datasets in Large Language Model (LLM) pre-training or post-training data. Contamination can lead to inflated scores on benchmarks, compromising evaluation results and making it difficult to determine the capabilities of models. In this work, we study the contamination of popular multilingual benchmarks in LLMs that support multiple languages. We use the Black Box test to determine whether $7$ frequently used multilingual benchmarks are contaminated in $7$ popular open and closed LLMs and find that almost all models show signs of being contaminated with almost all the benchmarks we test. Our findings can help the community determine the best set of benchmarks to use for multilingual evaluation.