Chart-to-Experience: Benchmarking Multimodal LLMs for Predicting Experiential Impact of Charts
机构 * KAIST（韩国科学技术院） ； Adobe Research（Adobe研究院）
Comments This paper has been accepted to IEEE PacificVis 2025
高校专区
Korea Advanced Institute of Science and Technology（韩国科学技术院）
Improving Chemical Understanding of LLMs via SMILES Parsing
机构 * KAIST（韩国科学技术院） ； Yonsei University（延世大学）
Dialogue in Resonance: An Interactive Music Piece for Piano and Real-Time Automatic Transcription System
机构 * Graduate School of Culture Technology, KAIST（文化科技研究生院，韩国科学技术院）
L2RDaS: Synthesizing 4D Radar Tensors for Model Generalization via Dataset Expansion
机构 * CCS Graduate School of Mobility, Korea Advanced Institute of Science and Technology（移动研究学院，韩国科学技术院）
Comments 9 pages, 3 figures, Arxiv preprint
SafeRoute: Adaptive Model Selection for Efficient and Accurate Safety Guardrails in Large Language Models
机构 * KAIST（韩国科学技术院） ； Technische Hochschule Nürnberg Georg Simon Ohm（图林根工业大学诺伊堡分校）
Comments ACL 2025 findings
Maximizing Discrimination Capability of Knowledge Distillation with Energy Function
机构 * KAIST（韩国科学技术院）
Comments 12 pages, 7 figures
Journal ref Knowl.-Based Syst. 296 (2024) 111911
Learning-based Autonomous Oversteer Control and Collision Avoidance
机构 * CCS Graduate School of Mobility, Korea Advanced Institute of Science and Technology（移动研究学院、韩国科学技术院）
Leveraging Generative AI Models to Explore Human Identity
机构 * Graduate School of Culture Technology, KAIST（韩国科学技术院文化科技研究生院） ； Samsung Advanced Institute of Technology (SAIT)（三星先进技术研究所）
Comments Accepted to ISEA 2025
Design of a 3-DOF Hopping Robot with an Optimized Gearbox: An Intermediate Platform Toward Bipedal Robots
机构 * Humanoid Robot Research Center, School of Mechanical, Aerospace & Systems Engineering, Department of Mechanical Engineering, Korea Advanced Institute of Science and Technology (KAIST)（人形机器人研究中心、机械航空航天与系统工程学院、机械工程系、韩国科学技术院）
EndoForce: Development of an Intuitive Axial Force Measurement Device for Endoscopic Robotic Systems
机构 * Department of Mechanical Engineering, Myongji University（Myongji大学机械工程系） ； R&D Department, Roen Surgical, Inc.（Roen Surgical公司研发部） ； Department of Mechnical Engineering, Korea Advanced Institute of Science and Technology（韩国科学技术院机械工程系） ； School of Mechatronics Engineering, Korea University of Technology and Education（韩国技术教育大学机电工程学院）
GeFL: Model-Agnostic Federated Learning with Generative Models
机构 * Samsung Electronics（三星电子） ； University of Texas at Austin（德克萨斯大学奥斯汀分校） ； Korea Advanced Institute of Science and Technology（韩国科学技术院）
Robustness-Reinforced Knowledge Distillation with Correlation Distance and Network Pruning
机构 * School of Electrical Engineering, Korea Advanced Institute of Science and Technology (KAIST)（电气工程学院，韩国科学技术院）
Comments 13 pages, 7 figures
Journal ref IEEE Trans. Knowl. Data Eng. 36 (2024) 9163-9175
Sample Efficient Reinforcement Learning via Large Vision Language Model Distillation
机构 * Robotics Program KAIST（韩国釜山科学技术院机器人计划） ； Electrical Engineering KAIST（韩国釜山科学技术院电子工程）
Comments 5 pages, ICASSP 2025. The first two authors are equally contributed
Journal ref ICASSP 2025 - 2025 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)
Latent Action Pretraining from Videos
机构 * KAIST（韩国科学技术院） ； University of Washington（华盛顿大学） ； Microsoft Research（微软研究院） ； NVIDIA（英伟达） ； Allen Institute for AI（人工智能算法研究所）
Comments ICLR 2025 Website: https://latentactionpretraining.github.io
"There Is No Such Thing as a Dumb Question," But There Are Good Ones
机构 * Interdisciplinary Program in Cognitive Science, Seoul National University（认知科学跨学科项目，首尔国立大学） ； Department of Brain and Cognitive Sciences, KAIST（脑科学与认知科学系，韩国科学技术院） ； Department of Physical Education, Dongguk University（体育系，东国大学）
Comments 8 pages, 4 figures and 4 tables. This work has been accepted for presentation as a poster with full paper publication at CogSci 2025. This is the final submission
Test-Time Augmentation for Pose-invariant Face Recognition
机构 * Korea Advanced Institute of Science and Technology（韩国科学技术院）
A Novel 6-axis Force/Torque Sensor Using Inductance Sensors
机构 * KAIST(Korea Advanced Institute of Science and Technology which is in the Daehak-Ro 291, Daejeon, South Korea（韩国科学技术院）
Comments 10 pages, 8 figures
VIMPPI: Enhancing Model Predictive Path Integral Control with Variational Integration for Underactuated Systems
机构 * Interactive Robotic Systems Laboratory, Korea Advanced Institute of Science and Technology (KAIST)（韩国科学技术院交互机器人系统实验室） ； Institute of Robotics and Computer Vision, Innopolis University（伊诺波利斯大学机器人与计算机视觉研究所）
Oaken: Fast and Efficient LLM Serving with Online-Offline Hybrid KV Cache Quantization
机构 * KAIST（韩国科学技术院） ； HyperAccel（HyperAccel公司）
Comments 16 pages, 14 figures, and 4 tables
Learning Advanced Self-Attention for Linear Transformers in the Singular Value Domain
机构 * Korea Advanced Institute of Science and Technology (KAIST)（韩国科学技术院） ； Yonsei University（延世大学）
Comments IJCAI25 Accepted
SHAP-based Explanations are Sensitive to Feature Representation
机构 * KAIST（韩国科学技术院） ； New York University（纽约大学）
Comments Accepted to ACM FAccT 2025
Reinforcement Learning-based Fault-Tolerant Control for Quadrotor with Online Transformer Adaptation
机构 * Agency for Defense Development（国防发展局） ； Korea Advanced Institute of Science and Technology（韩国先进科学技术研究院）
Comments Accpted at the 2025 IEEE International Conference on Robotics & Automation (ICRA) Workshop: Robots in the Wild
LightNobel: Improving Sequence Length Limitation in Protein Structure Prediction Model via Adaptive Activation Quantization
机构 * KAIST（韩国科学技术院）
Comments To appear in the Proceedings of the 52nd IEEE/ACM International Symposium on Computer Architecture (ISCA 2025)
Hearing and Seeing Through CLIP: A Framework for Self-Supervised Sound Source Localization
机构 * School of Electrical Engineering, KAIST, South Korea（韩国延世大学电气工程学院） ； ETRI, South Korea（韩国电子技术研究院）
Comments Journal Extension of WACV 2024 paper (arXiv:2311.04066). Code is available at ACL-SSL" target="_blank" rel="noopener">https://github.com/swimmiing/ACL-SSL
WorldCuisines: A Massive-Scale Benchmark for Multilingual and Multicultural Visual Question Answering on Global Cuisines
机构 * Capital One ； SEACrowd ； NAIST ； ITB ； UofT ； KAIST ； JAIST ； Cardiff University ； Independent ； HKUST ； SMU ； Masakhane ； University of Lagos ； HK PolyU ； Boston University ； Columbia University ； MBZUAI ； Monash University ； UCSB ； Tokyo Tech ； AI Singapore ； NICT ； Cohere ； McGill ； MILA ； Canada CIFAR AI Chair ； Ontario Tech
Comments Best Theme Paper at NAACL 2025
ExploreSelf: Fostering User-driven Exploration and Reflection on Personal Challenges with Adaptive Guidance by Large Language Models
机构 * KAIST（韩国科学技术院） ； NAVER Cloud（NAVER云） ； Northwestern University（西北大学） ； Georgia Institute of Technology（佐治亚理工学院） ； NAVER AI Lab（NAVER AI实验室）
Comments 17 pages excluding reference and appendix. Accepted at ACM CHI 2025. https://naver-ai.github.io/exploreself
Journal ref CHI '25: Proceedings of the 2025 CHI Conference on Human Factors in Computing Systems
AACessTalk: Fostering Communication between Minimally Verbal Autistic Children and Parents with Contextual Guidance and Card Recommendation
机构 * KAIST（韩国国立信息通信大学） ； NAVER Cloud（NAVER云） ； Dodakim Child Development Center（多达金儿童发展中心） ； NAVER AI Lab（NAVER AI实验室）
Comments 21 pages excluding reference. Accepted at ACM CHI 2025. https://naver-ai.github.io/aacesstalk/
Journal ref CHI '25: Proceedings of the 2025 CHI Conference on Human Factors in Computing Systems
TrajEvo: Designing Trajectory Prediction Heuristics via LLM-driven Evolution
机构 * KAIST（韩国科学技术院） ； OMELET ； UC Riverside（加州大学河滨分校）
$\texttt{SPIN}$: distilling $\texttt{Skill-RRT}$ for long-horizon prehensile and non-prehensile manipulation
机构 * Korea Advanced Institute of Science and Technology (KAIST)（韩国科学技术院）
Comments Project website: https://sites.google.com/view/skill-rrt
MARIOH: Multiplicity-Aware Hypergraph Reconstruction
机构 * Kim Jaechul Graduate School of AI, KAIST, Seoul, Republic of Korea（韩国科学技术院（KAIST）人工智能研究生院） ； GraphAI, Daejeon, Republic of Korea（韩国大田GraphAI）
Comments to be published in the 41st IEEE International Conference on Data Engineering (ICDE '25)
