Asankhaya Sharma

codelion

http://asankhaya.github.io/

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

liked a dataset about 10 hours ago

cais/hle

liked a model about 24 hours ago

LiquidAI/LFM2-350M

liked a model 5 days ago

Qwen/Qwen3-4B-Thinking-2507

View all activity

Organizations

upvoted a collection 15 days ago

Mem-Agent

Collection

Small sized agents from Dria trained on interacting with an obsidian-like memory system using python tools. Trained on Qwen3-4B-Thinking-2507. • 4 items • Updated 25 days ago • 3

upvoted a paper 17 days ago

BeyondWeb: Lessons from Scaling Synthetic Data for Trillion-scale Pretraining

Paper • 2508.10975 • Published Aug 14 • 59

upvoted an article 19 days ago

Article

mem-agent: Persistent, Human Readable Memory Agent Trained with Online RL

and 1 other •

19 days ago

• 24

upvoted a collection about 1 month ago

Nemotron-Pre-Training-Dataset

Collection

7 items • Updated 4 days ago • 34

upvoted an article about 2 months ago

Article

Building Enterprise-Ready Text Classifiers in Minutes with Adaptive Learning

•

Aug 9

• 12

upvoted 2 papers about 2 months ago

On the Generalization of SFT: A Reinforcement Learning Perspective with Reward Rectification

Paper • 2508.05629 • Published Aug 7 • 176

R-Zero: Self-Evolving Reasoning LLM from Zero Data

Paper • 2508.05004 • Published Aug 7 • 126

upvoted 2 articles about 2 months ago

Article

Towards Open Evolutionary Agents

and 1 other •

Aug 4

• 18

Article

Unsupervised Model Improvement via Internal Coherence Maximization: Outperforming Human-Supervised Methods Through Self-Elicitation

•

Aug 3

• 7

upvoted a collection 2 months ago

GLM-4.5

Collection

GLM-4.5: An open-source large language model designed for intelligent agents by Z.ai • 11 items • Updated Aug 11 • 241

upvoted 2 papers 2 months ago

Qwen3 Technical Report

Paper • 2505.09388 • Published May 14 • 294

Deep Researcher with Test-Time Diffusion

Paper • 2507.16075 • Published Jul 21 • 64

upvoted an article 2 months ago

Article

Understanding Model Reasoning Through Thought Anchors: A Comparative Study of Qwen3 and DeepSeek-R1

•

Jul 23

• 4

upvoted 2 collections 2 months ago

Ellora

Collection

Ellora: Enhancing LLMs with LoRA - Standardized Recipes for Capability Enhancement • 10 items • Updated Aug 3 • 2

Internal Coherence Maximization

Collection

Internal Coherence Maximization (ICM): A Label-Free, Unsupervised Training Framework for LLMs • 7 items • Updated Aug 3 • 2

upvoted a collection 3 months ago

Pre-training Dataset Samples

Collection

A collection of pre-training datasets samples of sizes 10M, 100M and 1B tokens. Ideal for use in quick experimentation and ablations. • 12 items • Updated 22 days ago • 4

upvoted 2 articles 3 months ago

Article

Automated Discovery of High-Performance GPU Kernels with OpenEvolve

•

Jun 27

• 22

Article

Adaptive Classifier: Dynamic Text Classification with Continuous Learning

•

Jun 20

• 17

upvoted 2 papers 3 months ago

ProtoReasoning: Prototypes as the Foundation for Generalizable Reasoning in LLMs

Paper • 2506.15211 • Published Jun 18 • 36

Reinforcement Learning with Verifiable Rewards Implicitly Incentivizes Correct Reasoning in Base LLMs

Paper • 2506.14245 • Published Jun 17 • 42

Asankhaya Sharma

AI & ML interests

Recent Activity

Organizations

codelion's activity

mem-agent: Persistent, Human Readable Memory Agent Trained with Online RL

Building Enterprise-Ready Text Classifiers in Minutes with Adaptive Learning

Towards Open Evolutionary Agents

Unsupervised Model Improvement via Internal Coherence Maximization: Outperforming Human-Supervised Methods Through Self-Elicitation

Understanding Model Reasoning Through Thought Anchors: A Comparative Study of Qwen3 and DeepSeek-R1

Automated Discovery of High-Performance GPU Kernels with OpenEvolve

Adaptive Classifier: Dynamic Text Classification with Continuous Learning