A high-throughput and memory-efficient inference and serving engine for LLMs

Finetune Llama 3.3, Mistral, Phi, Qwen 2.5 & Gemma LLMs 2-5x faster with 70% less memory

Efficient Triton Kernels for LLM Training

SEEN: Structured Event Enhancement Network for Explainable Need Detection of Information Recall Assistance

Learning to Generate Explanation from e-Hospital Services for Medical Suggestion

Contrastively learning participant representations per round in thread-based debates.

Analysis Model of Discourse Relations within a Document(AMDRD)