In a Training Loop 🔄
codelion
·
AI & ML interests
Creator of OptiLLM, OpenEvolve, Adaptive Classifier, and Ellora. Pioneering a new category in AI infrastructure: inference-time compute for LLMs.
Recent Activity
Organizations
-
-
-
-
-
-
-
-
-
-
-
published an
article about 1 month ago view article Reverse Engineering a $500M Mystery: From HashHop to Memory-Augmented Language Models
view article The Optimal Architecture for Small Language Models
view article Ellora: Enhancing LLMs with LoRA - Standardized Recipes for Capability Enhancement
view article The 1 Billion Token Challenge: Finding the Perfect Pre-training Mix
view article Building Enterprise-Ready Text Classifiers in Minutes with Adaptive Learning
view article Unsupervised Model Improvement via Internal Coherence Maximization: Outperforming Human-Supervised Methods Through Self-Elicitation
view article Understanding Model Reasoning Through Thought Anchors: A Comparative Study of Qwen3 and DeepSeek-R1
view article Automated Discovery of High-Performance GPU Kernels with OpenEvolve
view article Adaptive Classifier: Dynamic Text Classification with Continuous Learning
view article System Prompt Learning: Teaching LLMs to Learn Problem-Solving Strategies from Experience
view article AutoThink: Adaptive Reasoning for Large Language Models
view article OpenEvolve: An Open Source Implementation of Google DeepMind's AlphaEvolve
view article Introducing Pivotal Token Search (PTS): Targeting Critical Decision Points in LLM Training