PEMT: Multi-Task Correlation Guided Mixture-of-Experts Enables Parameter-Efficient Transfer Learningby Zhisheng Lin, Han Fu, Chenghao Liu,…
Unsupervised Domain Adaptation within Deep Foundation Latent Spacesby Dmitry Kangin, Plamen AngelovFirst submitted to arxiv…
Prompting a Pretrained Transformer Can Be a Universal Approximatorby Aleksandar Petrov, Philip H.S. Torr, Adel…
Fine-Tuning Enhances Existing Mechanisms: A Case Study on Entity Trackingby Nikhil Prakash, Tamar Rott Shaham,…
Visual Hallucinations of Multi-modal Large Language Modelsby Wen Huang, Hongbin Liu, Minxin Guo, Neil Zhenqiang…
Learning an Actionable Discrete Diffusion Policy via Large-Scale Actionless Video Pre-Trainingby Haoran He, Chenjia Bai,…
Word-Sequence Entropy: Towards Uncertainty Estimation in Free-Form Medical Question Answering Applications and Beyondby Zhiyuan Wang,…
HyperFast: Instant Classification for Tabular Databy David Bonet, Daniel Mas Montserrat, Xavier Giró-i-Nieto, Alexander G.…
Simple and Effective Transfer Learning for Neuro-Symbolic Integrationby Alessandro Daniele, Tommaso Campari, Sagar Malhotra, Luciano…
Contextual Molecule Representation Learning from Chemical Reaction Knowledgeby Han Tang, Shikun Feng, Bicheng Lin, Yuyan…