kim modal

kimmodal

6 13

kimmodal

AI & ML interests

None yet

Recent Activity

upvoted an article 1 day ago

MOPD: how frontier labs merge RL experts into one model and how to do it using TRL

liked a Space 9 days ago

witness-archive/alignment-hub

upvoted an article 18 days ago

Granite 4.2 LLMs: How They're Built

View all activity

Organizations

None yet

liked a Space 9 days ago

Alignment Hub

📚

Explore AI misalignment incident reports

liked 4 Spaces 3 months ago

QED-Nano: Teaching a Tiny Model to Prove Hard Theorems

📝

Who needs 1T parameters? Olympiad proofs with a 4B model

Don't Train the Model, Evolve the Harness

🌿

Evolving an agent's harness, not its model, on Harvey's LAB

The ultimate guide to RL environments: building and scaling them in the LLM era

📝

277

Building and scaling RL environments for LLM training

RL Environments 101 — Slides

🎞

Explore RL Environments 101 slide presentation

liked a dataset 6 months ago

SamsungSDS-Research/SDS-KoPub-VDR-Benchmark

Viewer • Updated Nov 11, 2025 • 33.3k • 675 • 15

liked a model 7 months ago

infly/Infinity-Parser2-Pro

Image-Text-to-Text • 35B • Updated Jul 20 • 2.24k • 105

liked a Space 7 months ago

Smol Training Playbook - Table of Contents

📚

liked 2 datasets 7 months ago

opendatalab/OmniDocBench

Viewer • Updated Jun 26 • 1.66k • 64.8k • 121

ling99/OCRBench_v2

Viewer • Updated Feb 24, 2025 • 10k • 3.58k • 20

liked a Space 7 months ago

The Smol Training Playbook

📚

3.35k

The secrets to building world-class LLMs