SAPIEN environment integrates PhysX and PartNet-Mobility, enabling detailed part-level interaction and perception tasks for household robots, advancing realism and diversity.

Fanbo Xiang, Yuzhe Qin, Kaichun Mo et al.

Proposes ELI framework for two-wave experiments under network interference, optimizing variance for precise causal inference.

Davide Viviano

This study evaluates BERT and RoBERTa calibration, showing temperature scaling and label smoothing significantly improve in- and out-of-domain calibration.

Shrey Desai, Greg Durrett

Axial-DeepLab uses axial-attention to outperform existing models on ImageNet, improving COCO test set by 2.8% PQ.

Huiyu Wang, Yukun Zhu, Bradley Green et al.

TREC 2019 deep learning track demonstrates that large-scale labeled data enables deep models like BERT to outperform traditional IR methods by up to 29.4%.

Nick Craswell, Bhaskar Mitra, Emine Yilmaz et al.

Introducing clDice, a topology-preserving loss for tubular structure segmentation, improving connectivity and graph similarity in 2D/3D data.

Suprosanna Shit, Johannes C. Paetzold, Anjany Sekuboyina et al.

GTI leverages trajectory intersections with GMM and cVAE to enable robots to generalize long-horizon tasks from few demonstrations, outperforming BC and GCBC.

Ajay Mandlekar, Danfei Xu, Roberto Martín-Martín et al.

Proposes a reinforcement learning curriculum framework using DAGs, classifies methods, emphasizes task sequencing and transfer mechanisms.

Sanmit Narvekar, Bei Peng, Matteo Leonetti et al.

Proposed a joint deconvolution and deep learning framework for UDC image restoration, achieving PSNR 28.5dB and 20fps real-time performance.

Yuqian Zhou, David Ren, Neil Emerton et al.

Automatic Curriculum Learning (ACL) optimizes sample efficiency and generalization in deep reinforcement learning.

Rémy Portelas, Cédric Colas, Lilian Weng et al.

Convolutional Occupancy Networks (CON) combine convolutional encoders with implicit occupancy decoders, achieving high-precision 3D reconstruction of complex scenes.

Songyou Peng, Michael Niemeyer, Lars Mescheder et al.

PathVQA employs semi-automated pipeline to create a dataset with 4,998 images and 32,799 QA pairs, advancing medical VQA research.

Xuehai He, Yichen Zhang, Luntian Mou et al.

StyleGAN2 distillation method enables feed-forward image manipulation with quality comparable to StyleGAN2 backpropagation.

Yuri Viazovetskyi, Vladimir Ivashkin, Evgeny Kashin

AutoML-Zero uses evolutionary search to discover complete ML algorithms from basic math operations, surpassing traditional AutoML limitations.

Esteban Real, Chen Liang, David R. So et al.

Proposes an adaptive task allocation framework using real-time performance feedback to update robot capability parameters without environmental models.

Yousef Emam, Siddharth Mayya, Gennaro Notomista et al.

Proposes ASAPP, an asynchronous distributed pose graph optimization algorithm, enhancing multi-robot SLAM robustness against communication delays.

Yulun Tian, Alec Koppel, Amrit Singh Bedi et al.

HPIPM offers a high-performance quadratic programming framework, significantly improving speed and robustness for model predictive control.

Gianluca Frison, Moritz Diehl

Proposes Fast-Slow GP-UCB for adversarially corrupted Bayesian optimization, achieving regret bounds that adapt to corruption level.

Ilija Bogunovic, Andreas Krause, Jonathan Scarlett

Proposes multi-microphone complex spectral mapping with DNN for dereverberation, combined with beamforming and post-filtering, achieving significant improvements.

Zhong-Qiu Wang, DeLiang Wang

LAMP employs laser-based multi-robot SLAM with scan-to-submap matching and ICM outlier rejection, achieving sub-meter accuracy in large underground environments.

Kamak Ebadi, Yun Chang, Matteo Palieri et al.