1 citations · 1 across the 6 of their papers we have counts for
6 papers
Exposure Bias Can Alleviate Itself via Directional and Frequency Rectification in Flow Matching
Guanbo Huang, Jingjia Mao, Fanding Huang +9
Flow Matching (FM) has achieved remarkable generative performance, yet it suffers from exposure bias due to discrepancies between training and inference. Existing mitigation strate…
ReflexFlow: Rethinking Learning Objective for Exposure Bias Alleviation in Flow Matching
Guanbo Huang, Jingjia Mao, Fanding Huang +8
Despite tremendous recent progress, Flow Matching methods still suffer from exposure bias due to discrepancies in training and inference. This paper investigates the root causes of…
Ask Question First for Enhancing Lifelong Language Learning
Han Wang, Ruiliu Fu, Xuejun Zhang +2
Lifelong language learning aims to stream learning NLP tasks while retaining knowledge of previous tasks. Previous works based on the language model and following data-free constra…
RVAE-LAMOL: Residual Variational Autoencoder to Enhance Lifelong Language Learning
Han Wang, Ruiliu Fu, Xuejun Zhang +1
Lifelong Language Learning (LLL) aims to train a neural network to learn a stream of NLP tasks while retaining knowledge from previous tasks. However, previous works which followed…
Decomposing Complex Questions Makes Multi-Hop QA Easier and More Interpretable
Ruiliu Fu, Han Wang, Xuejun Zhang +2
Multi-hop QA requires the machine to answer complex questions through finding multiple clues and reasoning, and provide explanatory evidence to demonstrate the machine reasoning pr…
Reminding the Incremental Language Model via Data-Free Self-Distillation
Han Wang, Ruiliu Fu, Chengzhang Li +3
Incremental language learning with pseudo-data can alleviate catastrophic forgetting in neural networks. However, to obtain better performance, former methods have higher demands f…