In a Training Loop 🔄

2 46 85

Piyush Maharana

catastropiyush

https://catastropiyush.github.io/

catastropiyush

AI & ML interests

LLMs for scientific data extraction, Solid State Hydrogen Storage,Machine Learning

Recent Activity

liked a dataset 7 days ago

jablonkagroup/euro_pmc_chemistry_abstracts

upvoted an article 10 days ago

Deriving the PPO Loss from First Principles

liked a Space 18 days ago

dlouapre/eiffel-tower-llama

View all activity

Organizations

upvoted an article 10 days ago

Article

Deriving the PPO Loss from First Principles

12 days ago

•

upvoted an article 28 days ago

Article

An Edge-First Generalized LLM LoRA Fine-Tuning Framework for Heterogeneous GPUs

Dec 1, 2025

•

upvoted an article about 1 month ago

Article

BERTs that chat: turn any BERT into a chatbot with dLLM

Nov 28, 2025

•

upvoted an article about 2 months ago

Article

TorchSim: A new PyTorch-based molecular dynamics engine

Oct 31, 2025

•

upvoted an article 2 months ago

Article

Supercharge your OCR Pipelines with Open Models

Oct 21, 2025

•

292

upvoted an article 4 months ago

Article

The Annotated Diffusion Model

Jun 7, 2022

•

307

upvoted an article 5 months ago

Article

Bringing Fusion Down to Earth: ML for Stellarator Optimization

Jul 2, 2025

•

upvoted an article 6 months ago

Article

Arc Virtual Cell Challenge: A Primer

Jul 18, 2025

•

upvoted a paper 6 months ago

Do Vision-Language Models Have Internal World Models? Towards an Atomic Evaluation

Paper • 2506.21876 • Published Jun 27, 2025 • 28

upvoted 2 articles 8 months ago

Article

Vision Language Models (Better, faster, stronger)

May 12, 2025

•

580

Article

Open-source DeepResearch – Freeing our search agents

Feb 4, 2025

•

1.31k

upvoted a collection 9 months ago

TxGemma Release

Collection

Collection of open models to accelerate the development of therapeutics. • 5 items • Updated Jul 10, 2025 • 66

upvoted a paper 9 months ago

HuatuoGPT-Vision, Towards Injecting Medical Visual Knowledge into Multimodal LLMs at Scale

Paper • 2406.19280 • Published Jun 27, 2024 • 63

upvoted an article 10 months ago

Article

What We Learned About LLM/VLMs in Healthcare AI Evaluation:

Nov 8, 2024

•

upvoted 5 articles 11 months ago

Article

How to generate text: using different decoding methods for language generation with Transformers

Mar 1, 2020

•

280

Article

The N Implementation Details of RLHF with PPO

Oct 24, 2023

•

Article

KV Caching Explained: Optimizing Transformer Inference Efficiency

Jan 30, 2025

•

210

Article

We now support VLMs in smolagents!

Jan 24, 2025

•

110

Article

SmolVLM Grows Smaller – Introducing the 256M & 500M Models!

Jan 23, 2025

•

189

upvoted an article 12 months ago

Article

Getting Started With Embeddings

Jun 23, 2022

•

101

Piyush Maharana

AI & ML interests

Recent Activity

Organizations

catastropiyush's activity

Deriving the PPO Loss from First Principles

**An Edge-First Generalized LLM LoRA Fine-Tuning Framework for Heterogeneous GPUs**

BERTs that chat: turn any BERT into a chatbot with dLLM

TorchSim: A new PyTorch-based molecular dynamics engine

Supercharge your OCR Pipelines with Open Models

The Annotated Diffusion Model

Bringing Fusion Down to Earth: ML for Stellarator Optimization

Arc Virtual Cell Challenge: A Primer

Vision Language Models (Better, faster, stronger)

Open-source DeepResearch – Freeing our search agents

What We Learned About LLM/VLMs in Healthcare AI Evaluation:

How to generate text: using different decoding methods for language generation with Transformers

The N Implementation Details of RLHF with PPO

KV Caching Explained: Optimizing Transformer Inference Efficiency

We now support VLMs in smolagents!

SmolVLM Grows Smaller – Introducing the 256M & 500M Models!

Getting Started With Embeddings

An Edge-First Generalized LLM LoRA Fine-Tuning Framework for Heterogeneous GPUs