Heegyu Kim PRO

heegyu's activity

• All
• Models
• Datasets
• Spaces
• Papers
• Collections
• Community
• Posts
• Upvotes
• Likes
• Articles

upvoted an article 7 days ago

view article

Article

Deriving the PPO Loss from First Principles

• 

10 days ago

•

31

upvoted an article about 2 months ago

view article

Article

Ultra-Long Sequence Parallelism: Ulysses + Ring-Attention Technical Principles and Implementation

• 

Sep 16, 2025

•

17

upvoted a paper 6 months ago

EXAONE 4.0: Unified Large Language Models Integrating Non-reasoning and Reasoning Modes

Paper • 2507.11407 • Published Jul 15, 2025 • 58

upvoted a paper 9 months ago

Quantization Hurts Reasoning? An Empirical Study on Quantized Reasoning Models

Paper • 2504.04823 • Published Apr 7, 2025 • 31

upvoted a paper 10 months ago

EgoSpeak: Learning When to Speak for Egocentric Conversational Agents in the Wild

Paper • 2502.14892 • Published Feb 17, 2025 • 6

upvoted an article 10 months ago

view article

Article

DABStep: Data Agent Benchmark for Multi-step Reasoning

• 
• 
• +4

Feb 4, 2025

•

122

upvoted a paper 10 months ago

Towards Fully-Automated Materials Discovery via Large-Scale Synthesis Dataset and Expert-Level LLM-as-a-Judge

Paper • 2502.16457 • Published Feb 23, 2025 • 11

upvoted a collection 11 months ago

🧠 Reasoning datasets

Collection

Datasets with reasoning traces for math and code released by the community • 24 items • Updated May 19, 2025 • 178

upvoted 7 collections about 1 year ago

EXAONE-3.5

Collection

EXAONE 3.5 language model series including instruction-tuned models of 2.4B, 7.8B, and 32B • 11 items • Updated Jul 7, 2025 • 119

Cosmos-Tokenizer

Collection

A suite of image and video tokenizers • 13 items • Updated 11 days ago • 43

Granite 3.0 Language Models

Collection

A series of language models trained by IBM licensed under Apache 2.0 license. We release both the base pretrained and instruct models. • 8 items • Updated Nov 17, 2025 • 100

Arch-Function

Collection

6 items • Updated Mar 28, 2025 • 9

LLM Safety Datasets

Collection

Korean safety, ethics dataset • 9 items • Updated Nov 23, 2024 • 3

En Ko Translate

Collection

영어 데이터셋을 한글로 번역한 데이터셋입니다. • 6 items • Updated Jul 15, 2025 • 2

Magpie Conversation Ko

Collection

Magpie 데이터셋 한국어 번역본 (@nayohan님 번역 모델 사용) • 10 items • Updated Nov 6, 2024 • 2

upvoted a paper about 1 year ago

MixEval: Deriving Wisdom of the Crowd from LLM Benchmark Mixtures

Paper • 2406.06565 • Published Jun 3, 2024 • 10

upvoted 4 collections over 1 year ago

3D

Collection

Stability AI's suite of models for 3D generation • 8 items • Updated Nov 14, 2025 • 50

4bit Instruct Models

Collection

18 items • Updated 11 days ago • 33

DeepSeek-Prover

Collection

DeepSeek-Prover-Series • 10 items • Updated Nov 27, 2025 • 60

Magpie-Qwen2 Datasets

Collection

Dataset built with Qwen2 72B and Qwen2 7B. • 6 items • Updated Jan 13, 2025 • 10