Showing cs.CLShow all
2 papers · 1 filter
cs.CL2026
When Perplexity Lies: Generation-Focused Distillation of Hybrid Sequence Models
Juan Gabriel Kostelec, Qinghai Guo
Converting a pretrained Transformer into a more efficient hybrid model through distillation offers a promising approach to reducing inference costs. However, achieving high-quality…
cs.CL2025
FlashEVA: Accelerating LLM inference via Efficient Attention
Juan Gabriel Kostelec, Qinghai Guo
Transformer models have revolutionized natural language processing, achieving state-of-the-art performance and demonstrating remarkable scalability. However, their memory demands,…