paper.dinq.meAI 与前沿科学顶会 · 学术新星雷达

更新于

返回新星榜

JK

Junmo Kim

NeurIPS 2026×6CVPR 2026多模态效率与系统检索与知识强化学习与后训练多语言与语言学

新星分资深

29

一作贡献0

核心/参与17

方向广度13

合作网络6

0 一作 · 7 篇 · NeurIPS 2026 / CVPR 2026

获取完整研究档案

首次获取该候选人的完整档案及 StartLux 评估共 20 积分；每位用户单独解锁，之后查看免费。主动刷新 20 积分。

联系方式

实时查找电子邮箱，费用不超过20积分；未找到邮箱不扣费。

2026 顶会论文

7

作者位次 #2/2NeurIPS 2026Poster· Deep Learning / Generative Models and Autoencoders

Module Specialization in Transformer Factual Recall under Correlated Fact Distributions

Dongjae Lee · Junmo Kim

检索与知识OpenReview

作者位次 #5/5NeurIPS 2026Poster· Applications / Language, Speech and Dialog

Continuous Audio Thinking for Large Audio Language Models

Gyojin Han · Dongjae Lee · Changho Choi · Jongsuk Kim · Junmo Kim

多模态OpenReview

作者位次 #4/5NeurIPS 2026Poster

HyperVAttention: Efficient Sparse Attention with Spatio-Temporal Clustering for Video Diffusion

Dongyeun Lee · Amir Zandieh · Vahab Mirrokni · Junmo Kim · Insu Han

多模态效率与系统OpenReview

作者位次 #5/5NeurIPS 2026Spotlight

Imitation Dominates Reinforcement: Direct In-Context RL Is Closer to ICL Than RL

Minchan Kwon · Seunghee Koh · Sunghyun Baek · Minsung Bae · Junmo Kim

强化学习与后训练OpenReview

作者位次 #4/4NeurIPS 2026Poster

Pygmalion Effect in Vision: Image-to-Clay Translation for Reflective Geometry Reconstruction

Gayoung Lee · Junho Kim · Jin-Hwa Kim · Junmo Kim

多模态多语言与语言学OpenReview

作者位次 #4/4NeurIPS 2026Poster

Before It Fades: Reinforcing Temporal Representations at Inference Time in VideoLLMs

Youngwoo Shin · Yusung Ro · Minseo Kim · Junmo Kim

可解释性多模态效率与系统OpenReview

作者位次 #2/2CVPR 2026Poster· Computer Vision & Pattern Recognition

Cross-Axis Feature Fusion with Joint-Wise Motion Difference Prediction for Text-Based 3D Human Motion Editing

Gyojin Han · Junmo Kim

多模态社会与人本OpenReview 项目页

常见合作者

DLDongjae Lee×2 GHGyojin Han×2 CCChangho Choi JKJongsuk Kim DLDongyeun Lee AZAmir Zandieh VMVahab Mirrokni IHInsu Han MKMinchan Kwon SKSeunghee Koh SBSunghyun Baek MBMinsung Bae