3 citations · 8 across the 18 of their papers we have counts for
22 papers
Shared Actors Need Not Share Critics: Effects of Value Mismatch in Parallel Reinforcement Learning
Zhenya Liu, Yang Meng, Zhuokai Zhao +2
When a single policy is trained in parallel across multiple environments of the same task, such as procedurally generated levels, randomized dynamics, or curricula, implementations…
Rethinking Transfer in Continual Learning: A Replay-Based Realisation
Yang Meng, Zhenya Liu, Zhuokai Zhao +1
Continual learning studies how deployed language models can continually acquire new tasks without expensive retraining from scratch. Existing methods, whether rehearsal-based (repl…
Accelerating PDE Surrogates via RL-Guided Mesh Optimization
Yang Meng, Ruoxi Jiang, Zhuokai Zhao +3
Deep surrogate models for parametric partial differential equations (PDEs) can deliver high-fidelity approximations but remain prohibitively data-hungry: training often requires th…
Scaling Agent Learning via Experience Synthesis
Zhaorun Chen, Zhuokai Zhao, Kai Zhang +15
While reinforcement learning (RL) can empower autonomous agents by enabling self-improvement through interaction, its practical adoption remains challenging due to costly rollouts,…
Boosting LLM Reasoning via Spontaneous Self-Correction
Xutong Zhao, Tengyu Xu, Xuewei Wang +11
While large language models (LLMs) have demonstrated remarkable success on a broad range of tasks, math reasoning remains a challenging one. One of the approaches for improving mat…
DISCO Balances the Scales: Adaptive Domain- and Difficulty-Aware Reinforcement Learning on Imbalanced Data
Yuhang Zhou, Jing Zhu, Shengyi Qian +7
Large Language Models (LLMs) are increasingly aligned with human preferences through Reinforcement Learning from Human Feedback (RLHF). Among RLHF methods, Group Relative Policy Op…
Transfer between Modalities with MetaQueries
Xichen Pan, Satya Narayan Shukla, Aashu Singh +9
Unified multimodal models aim to integrate understanding (text output) and generation (pixel output), but aligning these different modalities within a single architecture often dem…
Beyond Reward Hacking: Causal Rewards for Large Language Model Alignment
Chaoqi Wang, Zhuokai Zhao, Yibo Jiang +8
Recent advances in large language models (LLMs) have demonstrated significant progress in performing complex tasks. While Reinforcement Learning from Human Feedback (RLHF) has been…
Enhancing Vision-Language Model Reliability with Uncertainty-Guided Dropout Decoding
Yixiong Fang, Ziran Yang, Zhaorun Chen +2
Large vision-language models (LVLMs) excel at multimodal tasks but are prone to misinterpreting visual inputs, often resulting in hallucinations and unreliable outputs. We present…
Beyond Training: Dynamic Token Merging for Zero-Shot Video Understanding
Yiming Zhang, Zhuokai Zhao, Zhaorun Chen +3
Recent advancements in multimodal large language models (MLLMs) have opened new avenues for video understanding. However, achieving high fidelity in zero-shot video tasks remains c…
Quantifying Generalization Complexity for Large Language Models
Zhenting Qi, Hongyin Luo, Xuliang Huang +5
While large language models (LLMs) have shown exceptional capabilities in understanding complex queries and performing sophisticated tasks, their generalization abilities are often…
Preference Optimization with Multi-Sample Comparisons
Chaoqi Wang, Zhuokai Zhao, Chen Zhu +8
Recent advancements in generative models, particularly large language models (LLMs) and diffusion models, have been driven by extensive pretraining on large datasets followed by po…
MJ-Bench: Is Your Multimodal Reward Model Really a Good Judge for Text-to-Image Generation?
Zhaorun Chen, Yichao Du, Zichen Wen +16
While text-to-image models like DALLE-3 and Stable Diffusion are rapidly proliferating, they often encounter challenges such as hallucination, bias, and the production of unsafe, l…
RankCLIP: Ranking-Consistent Language-Image Pretraining
Yiming Zhang, Zhuokai Zhao, Zhaorun Chen +3
Self-supervised contrastive learning models, such as CLIP, have set new benchmarks for vision-language models in many downstream tasks. However, their dependency on rigid one-to-on…
HALC: Object Hallucination Reduction via Adaptive Focal-Contrast Decoding
Zhaorun Chen, Zhuokai Zhao, Hongyin Luo +3
While large vision-language models (LVLMs) have demonstrated impressive capabilities in interpreting multi-modal contexts, they invariably suffer from object hallucinations (OH). W…
AutoPRM: Automating Procedural Supervision for Multi-Step Reasoning via Controllable Question Decomposition
Zhaorun Chen, Zhuokai Zhao, Zhihong Zhu +4
Recent advancements in large language models (LLMs) have shown promise in multi-step reasoning tasks, yet their reliance on extensive manual labeling to provide procedural feedback…
Direct Acquisition Optimization for Low-Budget Active Learning
Zhuokai Zhao, Yibo Jiang, Yuxin Chen
Active Learning (AL) has gained prominence in integrating data-intensive machine learning (ML) models into domains with limited labeled data. However, its effectiveness diminishes…
Safe Reinforcement Learning via Hierarchical Adaptive Chance-Constraint Safeguards
Zhaorun Chen, Zhuokai Zhao, Tairan He +4
Ensuring safety in Reinforcement Learning (RL), typically framed as a Constrained Markov Decision Process (CMDP), is crucial for real-world exploration applications. Current approa…
RELAX: Reinforcement Learning Enabled 2D-LiDAR Autonomous System for Parsimonious UAVs
Guanlin Wu, Zhuokai Zhao, Yutao He
Unmanned Aerial Vehicles (UAVs) have become increasingly prominence in recent years, finding applications in surveillance, package delivery, among many others. Despite considerable…
Multimodal Guidance Network for Missing-Modality Inference in Content Moderation
Zhuokai Zhao, Harish Palani, Tianyi Liu +2
Multimodal deep learning, especially vision-language models, have gained significant traction in recent years, greatly improving performance on many downstream tasks, including con…
Evaluating Machine Learning Models with NERO: Non-Equivariance Revealed on Orbits
Zhuokai Zhao, Takumi Matsuzawa, William Irvine +2
Proper evaluations are crucial for better understanding, troubleshooting, interpreting model behaviors and further improving model performance. While using scalar-based error metri…
Breaking the Curse of Quality Saturation with User-Centric Ranking
Zhuokai Zhao, Yang Yang, Wenyu Wang +5
A key puzzle in search, ads, and recommendation is that the ranking model can only utilize a small portion of the vastly available user interaction data. As a result, increasing da…