62 citations · 113 across the 15 of their papers we have counts for
15 papers
Replay What Matters: Off-Policy Replay for Efficient LLM Reinforcement Unlearning
Zirui Pang, Chenlong Zhang, Haosheng Tan +3
LLM unlearning has emerged as a cost-effective alternative to full retraining for removing hazardous knowledge from pretrained models while preserving general utility. Recent RL-ba…
Incentivizing High-quality Participation From Federated Learning Agents
Jinlong Pang, Jiaheng Wei, Yifan Hua +2
Federated learning (FL) provides a promising paradigm for facilitating collaboration between multiple clients that jointly learn a global model without directly sharing their local…
Automatic Dataset Construction (ADC): Sample Collection, Data Curation, and Beyond
Minghao Liu, Zonglin Di, Jiaheng Wei +15
Large-scale data collection is essential for developing personalized training data, mitigating the shortage of training data, and fine-tuning specialized models. However, creating…
Measuring and Reducing LLM Hallucination without Gold-Standard Answers
Jiaheng Wei, Yuanshun Yao, Jean-Francois Ton +3
LLM hallucination, i.e. generating factually incorrect yet seemingly convincing answers, is currently a major threat to the trustworthiness and reliability of LLMs. The first step…
Human-Instruction-Free LLM Self-Alignment with Limited Samples
Hongyi Guo, Yuanshun Yao, Wei Shen +4
Aligning large language models (LLMs) with human values is a vital task for LLM practitioners. Current alignment techniques have several limitations: (1) requiring a large amount o…
Distributionally Robust Post-hoc Classifiers under Prior Shifts
Jiaheng Wei, Harikrishna Narasimhan, Ehsan Amid +3
The generalization ability of machine learning models degrades significantly when the test distribution shifts away from the training distribution. We investigate the problem of tr…
Human and AI Perceptual Differences in Image Classification Errors
Minghao Liu, Jiaheng Wei, Yang Liu +1
Artificial intelligence (AI) models for computer vision trained with supervised machine learning are assumed to solve classification tasks by imitating human behavior learned from…
Fairness Improves Learning from Noisily Labeled Long-Tailed Data
Jiaheng Wei, Zhaowei Zhu, Gang Niu +4
Both long-tailed and noisily labeled data frequently appear in real-world applications and impose significant challenges for learning. Most prior works treat either problem in an i…
Consensus on Dynamic Stochastic Block Models: Fast Convergence and Phase Transitions
Haoyu Wang, Jiaheng Wei, Zhenyuan Zhang
We introduce two models of consensus following a majority rule on time-evolving stochastic block models (SBM), in which the network evolution is Markovian or non-Markovian. Under t…
To Aggregate or Not? Learning with Separate Noisy Labels
Jiaheng Wei, Zhaowei Zhu, Tianyi Luo +3
The rawly collected training data often comes with separate noisy labels collected from multiple imperfect annotators (e.g., via crowdsourcing). A typical way of using these separa…
Learning with Noisy Labels Revisited: A Study Using Real-World Human Annotations
Jiaheng Wei, Zhaowei Zhu, Hao Cheng +3
Existing research on learning with noisy labels mainly focuses on synthetic label noise. Synthetic noise, though has clean structures which greatly enabled statistical analyses, of…
To Smooth or Not? When Label Smoothing Meets Noisy Labels
Jiaheng Wei, Hangyu Liu, Tongliang Liu +3
Label smoothing (LS) is an arising learning paradigm that uses the positively weighted average of both the hard training labels and uniformly distributed soft labels. It was shown…
DuelGAN: A Duel Between Two Discriminators Stabilizes the GAN Training
Jiaheng Wei, Minghao Liu, Jiahao Luo +3
In this paper, we introduce DuelGAN, a generative adversarial network (GAN) solution to improve the stability of the generated samples and to mitigate mode collapse. Built upon the…
When Optimizing -divergence is Robust with Label Noise
Jiaheng Wei, Yang Liu
We show when maximizing a properly defined -divergence measure with respect to a classifier's predictions and the supervised labels is robust with label noise. Leveraging its va…
Incentives for Federated Learning: a Hypothesis Elicitation Approach
Yang Liu, Jiaheng Wei
Federated learning provides a promising paradigm for collecting machine learning models from distributed data sources without compromising users' data privacy. The success of a cre…