Make the Pertinent Salient: Task-Relevant Reconstruction for Visual Control with Distractionsby Kyungmin Kim, JB Lanier,…
StatioCL: Contrastive Learning for Time Series via Non-Stationary and Temporal Contrastby Yu Wu, Ting Dang,…
On Discriminative Probabilistic Modeling for Self-Supervised Representation Learningby Bokun Wang, Yunwen Lei, Yiming Ying, Tianbao…
When Graph meets Multimodal: Benchmarking and Meditating on Multimodal Attributed Graphs Learningby Hao Yan, Chaozhuo…
SPA: 3D Spatial-Awareness Enables Effective Embodied Representationby Haoyi Zhu, Honghui Yang, Yating Wang, Jiange Yang,…
MolMix: A Simple Yet Effective Baseline for Multimodal Molecular Representation Learningby Andrei Manolache, Dragos Tantaru,…
Learning to Compress: Local Rank and Information Compression in Deep Neural Networksby Niket Patel, Ravid…
Compositional Entailment Learning for Hyperbolic Vision-Language Modelsby Avik Pal, Max van Spengler, Guido Maria D'Amely…
Causal Representation Learning in Temporal Data via Single-Parent Decodingby Philippe Brouillard, Sébastien Lachapelle, Julia Kaltenborn,…
MatMamba: A Matryoshka State Space Modelby Abhinav Shukla, Sai Vemprala, Aditya Kusupati, Ashish KapoorFirst submitted…