Dogun Kim

DiffusionDrive [CVPR 2025 Highlight]https://arxiv.org/abs/2411.15139 DiffusionDrive: Truncated Diffusion Model for End-to-End Autonomous DrivingRecently, the diffusion model has emerged as a powerful generative technique for robotic policy learning, capable of modeling multi-mode action distributions. Leveraging its capability for end-to-end autonomous driving is a promising direction. However, ..

Knowledge Distillation for Efficient Instance Semantic Segmentation with Transformers [CVPR 2024 Workshop]https://openaccess.thecvf.com/content/CVPR2024W/Vision4Ag/html/Li_Knowledge_Distillation_for_Efficient_Instance_Semantic_Segmentation_with_Transformers_CVPRW_2024_paper.html CVPR 2024 Open Access RepositoryKnowledge Distillation for Efficient Instance Semantic Segmentation with Transformers ..

DriveLM: Driving with Graph Visual Question Answering [ECCV 2024 Oral]https://arxiv.org/abs/2312.14150 DriveLM: Driving with Graph Visual Question AnsweringWe study how vision-language models (VLMs) trained on web-scale data can be integrated into end-to-end driving systems to boost generalization and enable interactivity with human users. While recent approaches adapt VLMs to driving via single..

Asynchronous Large Language Model Enhanced Planner for Autonomous Drivinghttps://arxiv.org/abs/2406.14556 Asynchronous Large Language Model Enhanced Planner for Autonomous DrivingDespite real-time planners exhibiting remarkable performance in autonomous driving, the growing exploration of Large Language Models (LLMs) has opened avenues for enhancing the interpretability and controllability of mo..

LLM+P: Empowering Large Language Modelswith Optimal Planning Proficiencyhttps://arxiv.org/abs/2304.11477 LLM+P: Empowering Large Language Models with Optimal Planning ProficiencyLarge language models (LLMs) have demonstrated remarkable zero-shot generalization abilities: state-of-the-art chatbots can provide plausible answers to many common questions that arise in daily life. However, so far, LL..

DriVLMe: Enhancing LLM-based Autonomous Driving Agents with Embodied and Social Experienceshttps://arxiv.org/abs/2406.03008 DriVLMe: Enhancing LLM-based Autonomous Driving Agents with Embodied and Social ExperiencesRecent advancements in foundation models (FMs) have unlocked new prospects in autonomous driving, yet the experimental settings of these studies are preliminary, over-simplified, and ..

Think2Drive: Efficient Reinforcement Learning by Thinking in Latent World Model for Quasi-Realistic Autonomous Driving in CARLA-v2https://arxiv.org/abs/2402.16720 Think2Drive: Efficient Reinforcement Learning by Thinking in Latent World Model for Quasi-Realistic Autonomous Driving (in CARLAReal-world autonomous driving (AD) especially urban driving involves many corner cases. The lately released..

Titans: Learning to Memorize at Test Timehttps://arxiv.org/abs/2501.00663 Titans: Learning to Memorize at Test TimeOver more than a decade there has been an extensive research effort on how to effectively utilize recurrent models and attention. While recurrent models aim to compress the data into a fixed-size memory (called hidden state), attention allows attending toarxiv.org 0. Abstract  RNN, ..

DriveGPT4: Interpretable End-to-end Autonomous Driving via Large Language Modelhttps://arxiv.org/abs/2310.01412 DriveGPT4: Interpretable End-to-end Autonomous Driving via Large Language ModelMultimodal large language models (MLLMs) have emerged as a prominent area of interest within the research community, given their proficiency in handling and reasoning with non-textual data, including images ..