PaperHub
影响力指数
97.99/100
前 0.1%
全站排名 #54
发表论文48 篇
平均评分5.7
年均产出16.0 篇/年

Jinwoo Shin

Full Professor@Korea Advanced Institute of Science & Technology·韩国·OpenReview
研究方向

Self/semi-supervised learning · Deep reinforcement learning · Generative adversarial networks · Out-of-distribution detection and generalization

6.0
23

HAMLET: Switch Your Vision-Language-Action Model into a History-Aware Policy

ICLR 2026Poster
通讯
5.5
17

Decoupled MeanFlow: Turning Flow Models into Flow Maps for Accelerated Sampling

ICLR 2026Poster
三作
5.5
25

Verifier-free Test-Time Sampling for Vision-Language-Action Models

ICLR 2026Poster
通讯
5.0
17

RedacBench: Can AI Erase Your Secrets?

ICLR 2026Poster
三作
5.0
22

SpatialBoost: Enhancing Visual Representation through Language-Guided Reasoning

ICLR 2026Rejected
通讯
4.7
17

ReGUIDE: Data Efficient GUI Grounding via Spatial Reasoning and Search

ICLR 2026Rejected
4.5
5

Rethinking Prompt Design for Inference-time Scaling in Text-to-Visual Generation

ICLR 2026Withdrawn
4.5
19

Contrastive Representation Regularization for Vision-Language-Action Models

ICLR 2026Rejected
通讯
4.0
21

Dual-Stream Diffusion for World-Model Augmented Vision-Language-Action Model

ICLR 2026Rejected
通讯
4.0
21

ContextVLA: Vision-Language-Action Model with Amortized Multi-Frame Context

ICLR 2026Rejected
通讯
4.0
19

RoboAlign: Reinforcement Learning for Action-Aligned Multimodal Large Language Models

ICLR 2026Rejected
通讯
3.0
5

INFUSER: Injecting Synthetic Failures for Self-Correcting Embodied Agents

ICLR 2026Withdrawn
通讯
2.8
6

Efficient LLM Collaboration via Planning

ICLR 2026Withdrawn
通讯
2.7
14

Scalable and Robust LLM Unlearning by Correcting Responses with Retrieved Exclusions

ICLR 2026Withdrawn
通讯
9.0
42

Representation Alignment for Generation: Training Diffusion Transformers Is Easier Than You Think

ICLR 2025Oral
8.7
18

Spread Preference Annotation: Direct Preference Judgment for Efficient LLM Alignment

ICLR 2025Oral
三作
7.8
21

MLE-STAR: Machine Learning Engineering Agent via Search and Targeted Refinement

NeurIPS 2025Poster
6.8
17

Track3R: Joint Point Map and Trajectory Prior for Spatiotemporal 3D Understanding

NeurIPS 2025Poster
二作
6.8
21

Self-Refining Language Model Anonymizers via Adversarial Distillation

NeurIPS 2025Poster
三作
6.7
22

Controllable Blur Data Augmentation Using 3D-Aware Motion Estimation

ICLR 2025Poster
通讯
6.5
34

Learning to Contextualize Web Pages for Enhanced Decision Making by LLM Agents

ICLR 2025Poster
6.4
35

Robot-R1: Reinforcement Learning for Enhanced Embodied Reasoning in Robotics

NeurIPS 2025Poster
通讯
6.2
35

Subtask-Aware Visual Reward Learning from Segmented Demonstrations

ICLR 2025Poster
6.1
12

Test-Time Adaptation with Binary Feedback

ICML 2025Poster
6.0
23

Binary-Feedback Active Test-Time Adaptation

ICLR 2025Rejected
6.0
32

DiffusionGuard: A Robust Defense Against Malicious Diffusion-based Image Editing

ICLR 2025Poster
6.0
22

Compress, Gather, and Recompute: REFORMing Long-Context Processing in Transformers

NeurIPS 2025Poster
5.5
20

Debiasing Online Preference Learning via Preference Feature Preservation

ICLR 2025Rejected
通讯
5.5
13

PRIME: Deep Imbalanced Regression with Proxies

ICML 2025Poster
通讯
4.9
15

Peri-LN: Revisiting Normalization Layer in the Transformer Architecture

ICML 2025Poster
4.8
11

ReVISE: Learning to Refine at Test-Time via Intrinsic Self-Verification

ICML 2025Poster
合作者 (20)