KAIST · embodied AI & robot learning韩国科学技术院 · 具身智能与机器人学习
7 curated papers, code, datasets and hardware from KAIST (3 with code), with one-line takeaways.7 个来自韩国科学技术院的精选论文、代码、数据集与硬件（3 个附代码），均配一句话要点。
Labs are matched from each entry's affiliation field by simple name rules (data/labs.yaml); joint work appears under every lab named. Only curated entries are listed, because daily arXiv papers carry no affiliations.机构由每个条目的机构字段按名称规则匹配（data/labs.yaml），合作工作会出现在所有署名机构下。仅列出精选条目，因为每日 arXiv 论文不含机构信息。
Reuses a pretrained video diffusion model to extract compact latent actions (1,200 to 4 tokens per frame) that act as a shared interface across humans, robots and environments, including zero-shot action transfer.复用预训练视频扩散模型提取紧凑潜动作（每帧 1,200 压到 4 个 token），作为人、机器人和不同环境之间的共享接口，并支持零示教动作迁移。
@misc{kim2026dreamx,
title = {{DreamX: Video Diffusion Model is a Universal Latent Action Learner}},
author = {Jisoo Kim and Chaehyun Kim and Junsu Kim and Jiho Park and Seongchan Kim and Eunbeen Hong and Soowon Son and Fu-En Yang and Seungryong Kim},
year = {2026},
eprint = {2611.05377},
archivePrefix = {arXiv},
primaryClass = {cs.CV},
url = {https://arxiv.org/abs/2611.05377},
}
Continual off-policy RL for VLAs — small actor/critic networks carry over between tasks, distilled on a few kept episodes to avoid forgetting; reports 31–68.5% less real-robot time on new assembly tasks.面向 VLA 的持续离策略强化学习：轻量 actor/critic 在任务间延续，用少量保留的旧任务轨迹做蒸馏以防遗忘。论文报告新的接触式装配任务所需真机训练时间减少 31–68.5%，旧任务几乎不遗忘。
@misc{kim2026accrue,
title = {{ACCRUE: Continual Off-Policy Reinforcement Learning for Robot Foundation Models}},
author = {Changyeon Kim and Kimin Lee and Yashraj Narang and Yijie Guo},
year = {2026},
eprint = {2610.18596},
archivePrefix = {arXiv},
primaryClass = {cs.RO},
url = {https://arxiv.org/abs/2610.18596},
}
DreamFree: a WAM that skips future-video generation. An encoder learns to predict a frozen video diffusion model's features from the current frame; actions come from those. Reports 11 Hz control with a 14B backbone.DreamFree：无需生成未来视频的世界动作模型。编码器从当前观测预测冻结视频扩散模型的表征，动作模块据此出动作；报告以 14B 视频扩散骨干实现 11Hz 实时闭环控制。
@misc{jo2026future,
title = {{Future Representation Distillation for Imagination-free World Action Models}},
author = {Jaehyeong Jo and Sangwon Jang and Sihyun Yu and Seonghyeon Ye and Yekyung Nah and Hojung Jung and Kimin Lee and Sung Ju Hwang},
year = {2026},
eprint = {2611.04527},
archivePrefix = {arXiv},
primaryClass = {cs.RO},
url = {https://arxiv.org/abs/2611.04527},
}
World Models世界模型world-action-modelvideo-diffusiondistillationreal-time
A humanoid arm whose force, speed and endurance targets come from human upper-limb measurements; reports a 20 kg peak payload (2.25x its own mass) and 20.6 m/s endpoint speed.人形手臂，其力量、速度和耐力指标取自人类上肢测量数据；报告峰值负载 20 公斤（自重的 2.25 倍），末端速度 20.6 米每秒。
@misc{kim2026raipal,
title = {{RAIPAL: A Powerful and Agile Humanoid Arm for Human Tasks}},
author = {Donggeon Kim and Hyungho Chris Choi and Donghyuk Choi and Jeongsoo Park and Choongin Lee and Chan Park and Jemin Hwangbo},
year = {2026},
eprint = {2611.04334},
archivePrefix = {arXiv},
primaryClass = {cs.RO},
url = {https://arxiv.org/abs/2611.04334},
}
FRED trains a world action model to predict better one-step future video features via self-distillation plus history context; reports large gains on long-horizon, out-of-domain and human-video-only tasks.FRED 通过自蒸馏并引入历史观测，让世界动作模型在单步去噪中预测出更好的未来视频表征；报告在长时程、域外环境以及只有人类视频示范的任务上提升明显。
@misc{jang2026self,
title = {{Self-Distilling Future Video Representations for World Action Models}},
author = {Huiwon Jang and Suhyeok Jang and Hyeonmo Kang and Seungku Kim and Jinwoo Shin and Sihyun Yu},
year = {2026},
eprint = {2610.30320},
archivePrefix = {arXiv},
primaryClass = {cs.RO},
url = {https://arxiv.org/abs/2610.30320},
}
World Models世界模型world-action-modelvideo-diffusionself-distillationhuman-to-robotreal
Paper论文 2024 · KAIST / UW / Microsoft Research / NVIDIA ICLR 2025
LAPA pretrains a VLA without action labels: it learns discrete latent actions between video frames, predicts them from image + text, then maps them to robot actions; reports beating OpenVLA on real tasks.LAPA 无需动作标注即可预训练 VLA：先从视频相邻帧间学习离散隐动作，再训练模型由图像和指令预测隐动作，最后用少量机器人数据映射到真实动作；报告在真机任务上优于 OpenVLA。
@misc{ye2024latent,
title = {{Latent Action Pretraining from Videos}},
author = {Seonghyeon Ye and Joel Jang and Byeongguk Jeon and Sejune Joo and Jianwei Yang and Baolin Peng and Ajay Mandlekar and Reuben Tan and Yu-Wei Chao and Bill Yuchen Lin and Lars Liden and Kimin Lee and Jianfeng Gao and Luke Zettlemoyer and Dieter Fox and Minjoon Seo},
year = {2024},
eprint = {2410.11758},
archivePrefix = {arXiv},
primaryClass = {cs.RO},
url = {https://arxiv.org/abs/2410.11758},
}
A blind quadruped policy that infers terrain properties implicitly from proprioception history with a context-aided estimator, enabling robust walking on stairs and rough ground without cameras.盲视四足策略：借助上下文辅助估计器从本体感觉历史中隐式推断地形特性，无需相机即可稳健地走楼梯和崎岖地面。