Blake Olson

M.S. Student, UT Austin

blakeolson@utexas.edu

Bio

I am a master's student at The University of Texas at Austin, focusing on the planning and reasoning abilities of large language models. Specifically, I am interested in reinforcement learning and post-training to improve reasoning ability.

My most recent work investigates the impact of curriculum learning on reasoning. We introduce E2H Reasoner (Easy2Hard), a method that improves post-training via cosine or gaussian scheduling. We compare supervised fine-tuning to reinforcement learning and measure their impact on trivial, easy, medium, and hard tasks.

Publications

2026

Curriculum Reinforcement Learning from Easy to Hard Tasks Improves LLM Reasoning

Shubham Parashar*, Shurui Gui*, Xiner Li*, Hongyi Ling, Sushil Vemuri, Blake Olson, Eric Li, Yu Zhang, James Caverlee, Dileep Kalathil, Shuiwang Ji

International Conference on Learning Representations (ICLR), 2026

Paper
2025

Complex LLM Planning via Automated Heuristics Discovery

Hongyi Ling*, Shubham Parashar*, Sambhav Khurana*, Blake Olson, Anwesha Basu, Gaurangi Sinha, Zhengzhong Tu, James Caverlee, Shuiwang Ji

arXiv preprint, 2025

Paper

Inference-Time Computations for LLM Reasoning and Planning: A Benchmark and Insights

Shubham Parashar*, Blake Olson*, Sambhav Khurana*, Eric Li*, Hongyi Ling, James Caverlee, Shuiwang Ji

arXiv preprint, 2025

Paper Code

Fragment and Geometry Aware Tokenization of Molecules for Structure-Based Drug Design Using Language Models

Cong Fu*, Xiner Li*, Blake Olson, Heng Ji, Shuiwang Ji

International Conference on Learning Representations (ICLR), 2025

Paper Code

(* indicates equal contribution)

Vitæ

Full Resume in PDF.

Blake Olson
blakeolson@utexas.edu