2 citations · 4 across the 12 of their papers we have counts for
13 papers
Dr. OPD: Learning What to Follow for Optimal On-Policy Distillation of Large Language Models
Zhenyu Wang, Tianze Wang, Linjun Zhang +1
On-policy distillation (OPD) trains a student on its own generated responses using dense, token-level supervision from a stronger teacher. Vanilla OPD treats all teacher signals eq…
CCAD: Compressed Global Feature Conditioned Anomaly Detection
Xiao Jin, Liang Diao, Qixin Xiao +4
Anomaly detection holds considerable industrial significance, especially in scenarios with limited anomalous data. Currently, reconstruction-based and unsupervised representation-b…
Unlocking the Essence of Beauty: Advanced Aesthetic Reasoning with Relative-Absolute Policy Optimization
Boyang Liu, Yifan Hu, Senjie Jin +5
Multimodal large language models (MLLMs) are well suited to image aesthetic assessment, as they can capture high-level aesthetic features leveraging their cross-modal understanding…
Entropy-Driven Pre-Tokenization for Byte-Pair Encoding
Yifan Hu, Frank Liang, Dachuan Zhao +4
Byte-Pair Encoding (BPE) has become a widely adopted subword tokenization method in modern language models due to its simplicity and strong empirical performance across downstream…
Reinforcement Learning with Continuous Actions Under Unmeasured Confounding
Yuhan Li, Eugene Han, Yifan Hu +4
This paper addresses the challenge of offline policy learning in reinforcement learning with continuous action spaces when unmeasured confounders are present. While most existing r…
Global Group Fairness in Federated Learning via Function Tracking
Yves Rychener, Daniel Kuhn, Yifan Hu
We investigate group fairness regularizers in federated learning, aiming to train a globally fair model in a distributed setting. Ensuring global fairness in distributed training p…
MPO: An Efficient Post-Processing Framework for Mixing Diverse Preference Alignment
Tianze Wang, Dongnan Gui, Yifan Hu +2
Reinforcement Learning from Human Feedback (RLHF) has shown promise in aligning large language models (LLMs). Yet its reliance on a singular reward model often overlooks the divers…
Efficiently Serving Large Multimodal Models Using EPD Disaggregation
Gursimran Singh, Xinglu Wang, Yifan Hu +9
Large Multimodal Models (LMMs) extend Large Language Models (LLMs) by handling diverse inputs such as images, audio, and video, but at the cost of adding a multimodal encoding stag…
Contextual Bilevel Reinforcement Learning for Incentive Alignment
Vinzenz Thoma, Barna Pasztor, Andreas Krause +2
The optimal policy in various real-world strategic decision-making problems depends both on the environmental configuration and exogenous events. For these settings, we introduce C…
Group Robust Preference Optimization in Reward-free RLHF
Shyam Sundhar Ramesh, Yifan Hu, Iason Chaimalas +4
Adapting large language models (LLMs) for specific tasks usually involves fine-tuning through reinforcement learning with human feedback (RLHF) on preference data. While these data…
Stochastic Optimization Algorithms for Instrumental Variable Regression with Streaming Data
Xuxing Chen, Abhishek Roy, Yifan Hu +1
We develop and analyze algorithms for instrumental variable regression by viewing the problem as a conditional stochastic optimization problem. In the context of least-squares inst…
Contextual Stochastic Bilevel Optimization
Yifan Hu, Jie Wang, Yao Xie +2
We introduce contextual stochastic bilevel optimization (CSBO) -- a stochastic bilevel optimization framework with the lower-level problem minimizing an expectation conditioned on…
Distributionally Robust Model-based Reinforcement Learning with Large State Spaces
Shyam Sundhar Ramesh, Pier Giuseppe Sessa, Yifan Hu +2
Three major challenges in reinforcement learning are the complex dynamical systems with large state spaces, the costly data acquisition processes, and the deviation of real-world d…