Lucas Bandarkar

lucasbandarkar's activity

• All
• Models
• Datasets
• Spaces
• Buckets
• Papers
• Collections
• Community
• Posts
• Upvotes
• Likes
• Articles

authored 6 papers 2 days ago

Cross-Lingual Alignment for Decoder-Only Models using MoE Routers

Paper • 2610.01921 • Published 8 days ago • 2

Multilinguality in Hybrid Attention LLMs

Paper • 2609.35378 • Published 12 days ago • 8

OmnilingualGAIA2: Evaluating the Multilingual Gap in Frontier AI Agents

Paper • 2608.08775 • Published Aug 9

Extracting Small Translation Specialists from LLMs by Aggressively Pruning Experts

Paper • 2605.28042 • Published May 27 • 1

Translation as a Scalable Proxy for Multilingual Evaluation

Paper • 2601.11778 • Published Jan 16 • 1

Multilingual Routing in Mixture-of-Experts

Paper • 2510.04694 • Published Oct 6, 2025 • 7

submitted 2 papers to Daily Papers 2 days ago

Cross-Lingual Alignment for Decoder-Only Models using MoE Routers

Paper • 2610.01921 • Published 8 days ago • 2

Multilinguality in Hybrid Attention LLMs

Paper • 2609.35378 • Published 12 days ago • 8

authored a paper almost 2 years ago

Layer Swapping for Zero-Shot Cross-Lingual Transfer in Large Language Models

Paper • 2410.01335 • Published Oct 2, 2024 • 5

authored a paper about 3 years ago

The Belebele Benchmark: a Parallel Reading Comprehension Dataset in 122 Language Variants

Paper • 2308.16884 • Published Aug 31, 2023 • 11