43 citations · 43 across the 12 of their papers we have counts for
16 papers
Cross-Lingual Alignment for Decoder-Only Models using MoE Routers
Lucas Bandarkar, Clark Peng, Ahmed Haj Ahmed +2
Cross-lingual contrastive learning has been a core component of multilingual encoder training, but the ability to explicitly align representations is not possible in decoder-only L…
Multilinguality in Hybrid Attention LLMs
Lucas Bandarkar, Junlin Hu, Chenyuan Yang +2
In response to the growing demand for long sequences in agentic and reasoning use cases, many state-of-the-art LLMs combine multiple variants of attention to mitigate the quadratic…
ESTS at WMT26: Routing-Informed Expert Pruning for Model Compression
Liu O. Martin, Lucas Bandarkar, Nanyun Peng
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 submi…
OmnilingualGAIA2: Evaluating the Multilingual Gap in Frontier AI Agents
Andrea Caciolai, Pere-Lluís Huguet Cabot, Chierh Cheng +11
Agentic benchmarks aim to measure how well AI agents plan, search, execute, and recover within realistic multi-tool environments, but they are almost exclusively in English. As AI…
Extracting Small Translation Specialists from LLMs by Aggressively Pruning Experts
Liu O. Martin, Lucas Bandarkar, Nanyun Peng
Modern large language models (LLMs) achieve state-of-the-art machine translation performance, but they do so as broad generalists largely trained for many tasks and capabilities un…
Knowledge Localization in Mixture-of-Experts LLMs Using Cross-Lingual Inconsistency
Lucas Bandarkar, Alan Ansell, Trevor Cohn
Modern LLMs continue to exhibit significant variance in behavior across languages, such as being able to recall factual information in some languages but not others. While typicall…
Large Reasoning Models Struggle to Transfer Parametric Knowledge Across Scripts
Lucas Bandarkar, Alan Ansell, Trevor Cohn
In this work, we analyze shortcomings in cross-lingual knowledge transfer in large, modern reasoning LLMs. We demonstrate that the perceived gap in knowledge transfer is primarily…
Translation as a Scalable Proxy for Multilingual Evaluation
Sheriff Issaka, Erick Rosas Gonzalez, Lieqi Liu +6
The rapid proliferation of LLMs has created a critical evaluation paradox: while LLMs claim multilingual proficiency, comprehensive non-machine-translated benchmarks exist for fewe…
Multilingual Routing in Mixture-of-Experts
Lucas Bandarkar, Chenyuan Yang, Mohsen Fayyaz +2
Mixture-of-Experts (MoE) architectures have become the key to scaling modern LLMs, yet little is understood about how their sparse routing dynamics respond to multilingual data. In…
IndiaWeatherBench: A Dataset and Benchmark for Data-Driven Regional Weather Forecasting over India
Tung Nguyen, Harkanwar Singh, Nilay Naharas +2
Regional weather forecasting is a critical problem for localized climate adaptation, disaster mitigation, and sustainable development. While machine learning has shown impressive p…
The Unreasonable Effectiveness of Model Merging for Cross-Lingual Transfer in LLMs
Lucas Bandarkar, Nanyun Peng
Large language models (LLMs) still struggle across tasks outside of high-resource languages. In this work, we investigate cross-lingual transfer to lower-resource languages where t…
SNaRe: Domain-aware Data Generation for Low-Resource Event Detection
Tanmay Parekh, Yuxuan Dong, Lucas Bandarkar +4
Event Detection (ED) -- the task of identifying event mentions from natural language text -- is critical for enabling reasoning in highly specialized domains such as biomedicine, l…
Layer Swapping for Zero-Shot Cross-Lingual Transfer in Large Language Models
Lucas Bandarkar, Benjamin Muller, Pritish Yuvraj +4
Model merging, such as model souping, is the practice of combining different models with the same architecture together without further training. In this work, we present a model m…
The Belebele Benchmark: a Parallel Reading Comprehension Dataset in 122 Language Variants
Lucas Bandarkar, Davis Liang, Benjamin Muller +7
We present Belebele, a multiple-choice machine reading comprehension (MRC) dataset spanning 122 language variants. Significantly expanding the language coverage of natural language…
Can Transformer Models Measure Coherence In Text? Re-Thinking the Shuffle Test
Philippe Laban, Luke Dai, Lucas Bandarkar +1
The Shuffle Test is the most common task to evaluate whether NLP models can measure coherence in text. Most recent work uses direct supervision on the task; we show that by simply…
News Headline Grouping as a Challenging NLU Task
Philippe Laban, Lucas Bandarkar, Marti A. Hearst
Recent progress in Natural Language Understanding (NLU) has seen the latest models outperform human performance on many standard tasks. These impressive results have led the commun…