Arunkumar Venkataramanan's picture

In a Training Loop 🔄

139 291

Arunkumar Venkataramanan

ArunkumarVR

·

https://arunkumarramanan.github.io

AI & ML interests

AGI Research: Reasoning, Safety & Alignment (Superalignment), Generative AI (GenAI), Multi-Modal Foundation Models (FMs), Large Language Models (LLMs), Transformers & Diffusion Models, Open LLM Training, Optimization & Finetuning, Serving & Inference

Recent Activity

liked a dataset about 2 hours ago

Skylion007/openwebtext

liked a model 3 days ago

tencent/WeDLM-8B-Instruct

liked a model 3 days ago

tencent/WeDLM-8B-Base

View all activity

Organizations

upvoted a paper 10 days ago

SmolLM2: When Smol Goes Big -- Data-Centric Training of a Small Language Model

Paper • 2502.02737 • Published Feb 4, 2025 • 252

upvoted a collection 14 days ago

Moonlight-A3B

Moonshot's Compute-efficient MoE LLM, first Scaling Up of Muon Optimizer • 3 items • Updated Nov 2, 2025 • 9

upvoted a paper 14 days ago

Kimi k1.5: Scaling Reinforcement Learning with LLMs

Paper • 2501.12599 • Published Jan 22, 2025 • 126

upvoted a collection 15 days ago

FunctionGemma

3 items • Updated 16 days ago • 30

upvoted an article 16 days ago

Article

The Open Evaluation Standard: Benchmarking NVIDIA Nemotron 3 Nano with NeMo Evaluator

17 days ago

•

35

upvoted 2 collections 18 days ago

Nemotron v3 Pre-Training

Large scale pre-training datasets used in the Nemotron family of models. • 11 items • Updated 11 days ago • 6

Common Pile v0.1

All resources related to Common Pile v0.1, an 8TB dataset of public domain and openly licensed text • 4 items • Updated Jun 6, 2025 • 39

upvoted an article 18 days ago

Article

Nemotron 3 Nano \- A new Standard for Efficient, Open, and Intelligent Agentic Models

19 days ago

•

104

upvoted a paper 18 days ago

ReFusion: A Diffusion Large Language Model with Parallel Autoregressive Decoding

Paper • 2512.13586 • Published 19 days ago • 87

upvoted 2 collections 19 days ago

Nemotron-Post-Training-v3

Collection of datasets used in the post-training phase of Nemotron Nano v3. • 7 items • Updated 11 days ago • 54

NVIDIA Nemotron v3

Open, Production-ready Enterprise Models • 6 items • Updated 4 days ago • 109

upvoted 2 collections 21 days ago

Nemotron-Pre-Training-Datasets

Large scale pre-training datasets used in the Nemotron family of models. • 11 items • Updated 11 days ago • 85

NeMo Gym

Collection of RL verifiable data for NeMo Gym • 13 items • Updated 11 days ago • 32

upvoted a collection 24 days ago

Devstral 2

A couple of agentic LLMs for software engineering tasks, excelling at using tools to explore codebases, edit multiple files, and power SWE Agents. • 3 items • Updated 25 days ago • 38

upvoted a collection 28 days ago

Essential-Web v1.0

10 items • Updated Jun 18, 2025 • 10

upvoted a paper 28 days ago

Rethinking Reflection in Pre-Training

Paper • 2504.04022 • Published Apr 5, 2025 • 80

upvoted an article 29 days ago

Article

We Got Claude to Fine-Tune an Open Source LLM

about 1 month ago

•

556

upvoted an article about 1 month ago

Article

Transformers v5: Simple model definitions powering the AI ecosystem

+2

Dec 1, 2025

•

263

upvoted 2 collections about 1 month ago

Mistral Large 3

A state-of-the-art, open-weight, general-purpose multimodal model with a granular Mixture-of-Experts architecture. • 4 items • Updated Dec 2, 2025 • 81

Ministral 3

A collection of edge models, with Base, Instruct and Reasoning variants, in 3 different sizes: 3B, 8B and 14B. All with vision capabilities. • 9 items • Updated Dec 2, 2025 • 135