DongGeon Lee

oneonlee

·

https://donggeon.github.io

AI & ML interests

Data-centric natural language processing, AI Safety & Alignment

Recent Activity

liked a dataset 2 days ago

jumplander/JL-Agentic-Behavior-10K

liked a model 5 days ago

meta-models/Muse-Glimmer-30B

liked a dataset 23 days ago

View all activity

Organizations

None yet

upvoted an article 25 days ago

Article

Security incident disclosure — July 2026

system

•

Jul 16

• 773

upvoted a collection about 2 months ago

OpenThinker-Agent

5 items • Updated Jun 11 • 11

upvoted a collection 4 months ago

Nemotron-Post-Training-v3

Collection of datasets used in the post-training phase of Nemotron Nano, Super, and Ultra v3. • 50 items • Updated 5 days ago • 195

upvoted an article 4 months ago

Article

How to Ground a Korean AI Agent in Real Demographics with Synthetic Personas

nvidia

•

Apr 21

• 27

upvoted a paper 5 months ago

Scaling Synthetic Data Creation with 1,000,000,000 Personas

Paper • 2406.20094 • Published Jun 28, 2024 • 107

upvoted a paper 7 months ago

What Users Leave Unsaid: Under-Specified Queries Limit Vision-Language Models

Paper • 2601.06165 • Published Jan 7 • 16

upvoted a collection 7 months ago

COMPASS

A Framework for Evaluating Organization-Specific Policy Alignment in LLMs • 5 items • Updated Jan 6 • 5

upvoted a paper 7 months ago

COMPASS: A Framework for Evaluating Organization-Specific Policy Alignment in LLMs

Paper • 2601.01836 • Published Jan 5 • 10

upvoted 2 collections 10 months ago

Qwen3Guard

7 items • Updated Dec 31, 2025 • 70

gpt-oss-safeguard

gpt-oss-safeguard-120b and gpt-oss-safeguard-20b are safety reasoning models built-upon gpt-oss • 2 items • Updated Oct 29, 2025 • 75

upvoted 2 papers 10 months ago

The Attacker Moves Second: Stronger Adaptive Attacks Bypass Defenses Against Llm Jailbreaks and Prompt Injections

Paper • 2510.09023 • Published Oct 10, 2025 • 11

Qwen3Guard Technical Report

Paper • 2510.14276 • Published Oct 16, 2025 • 16

upvoted a collection about 1 year ago

Mi:dm 2.0

KT's Korea-centric AI models - Mi:dm 2.0 • 2 items • Updated Sep 24, 2025 • 16

upvoted a paper about 1 year ago

Are Vision-Language Models Safe in the Wild? A Meme-Based Benchmark Study

Paper • 2505.15389 • Published May 21, 2025 • 8

upvoted 2 papers over 1 year ago

Lost in Cultural Translation: Do LLMs Struggle with Math Across Cultural Contexts?

Paper • 2503.18018 • Published Mar 23, 2025 • 7

Typed-RAG: Type-aware Multi-Aspect Decomposition for Non-Factoid Question Answering

Paper • 2503.15879 • Published Mar 20, 2025 • 6

upvoted 2 collections over 1 year ago

DeepSeek-R1

10 items • Updated Nov 27, 2025 • 862

Korean Math Dataset

한국어 수학 데이터 • 15 items • Updated 10 days ago • 13

upvoted a collection almost 2 years ago

BERDS

The BERDS Benchmark aims to measure retrieval diversity for questions that are opinionated or invite diverse perspectives. • 5 items • Updated Sep 25, 2024 • 2

upvoted a paper about 2 years ago

A Tale of Trust and Accuracy: Base vs. Instruct LLMs in RAG Systems

Paper • 2406.14972 • Published Jun 21, 2024 • 7