Publications
KRAFTON AI 는 자체 연구 뿐만 아니라 학계와 협업하여 연구 영역을 확장하고 있습니다.
*볼드체로 된 저자는 크래프톤 소속임
Filter
Online Agent-as-a-Judge: Situation-Generating Evaluation for Interactive Agents
Online Agent-as-a-Judge: Situation-Generating Evaluation for Interactive Agents
Pruning and Distilling Mixture-of-Experts into Dense Language Models
Pruning and Distilling Mixture-of-Experts into Dense Language Models
Uniform Spectral Growth under Factor-wise Muon Orthogonalization in Matrix Factorization and LoRA
Uniform Spectral Growth under Factor-wise Muon Orthogonalization in Matrix Factorization and LoRA
AMUSE: Anytime Muon with Stable Gradient Evaluation
AMUSE: Anytime Muon with Stable Gradient Evaluation
RoDAC: A Robust Data-centric Anti-Cheat Framework for Fair Online Competitive Gaming
RoDAC: A Robust Data-centric Anti-Cheat Framework for Fair Online Competitive Gaming
FlashSAC: Fast and Stable Off-Policy Reinforcement Learning for High-Dimensional Robot Control
FlashSAC: Fast and Stable Off-Policy Reinforcement Learning for High-Dimensional Robot Control
Identifiable Token Correspondence for World Models
Identifiable Token Correspondence for World Models
Fine-tuning a vision-language model for fracture-surface morphology recognition
Fine-tuning a vision-language model for fracture-surface morphology recognition
Convex Distance Operator Transport: Convex and Geometry-Preserving Formulation
Convex Distance Operator Transport: Convex and Geometry-Preserving Formulation
How to Correctly Report LLM-as-a-Judge Evaluations
