Research Highlights

---

"PLDR-LLMs Learn A Generalizable Tensor Operator That Can Replace Its Own Deep Neural Net At Inference." arXiv preprint arXiv:2502.13502 (2025) 

"PLDR-LLM: Large Language Model from Power Law Decoder Representations." arXiv preprint arXiv:2410.16703 (2024). 

"Power Law Graph Transformer for Machine Translation and Representation Learning." arXiv preprint arXiv:2107.02039 (2021). 

"Coulgat: An experiment on interpretability of graph attention networks." arXiv preprint arXiv:1912.08409 (2019).