42 citations · 57 across the 25 of their papers we have counts for
25 papers
Beyond Isolation: Unlocking Reinforcement Learning Component Synergy for Sample-Efficient Continuous Control
Qi Zhao, Guozheng Ma, Yilun Kong +9
Reinforcement learning systems are significantly more complex than other machine learning paradigms due to inherent properties, causing RL system design to jointly account for many…
GaitSpan: Growing Humanoid Locomotion from Walking to Running
Kwan-Yee Lin, Zilin Wang, Janelle J. Liu +1
A humanoid that can walk should not relearn locomotion from scratch to jog or run. Yet current approaches often obtain gait diversity by prescribing gait schedules, imitating motio…
Beyond One-Size-Fits-All: Diagnosis-Driven Online Reinforcement Learning with Offline Priors
Guozheng Ma, Lu Li, Zilin Wang +2
Online reinforcement learning (RL) agents increasingly depend on knowledge acquired offline to achieve practical efficiency. Originally studied in offline-to-online RL, this paradi…
Vision Harnessing Agent for Open Ad-hoc Segmentation
Zilin Wang, Stella X. Yu
Segmentation has become easy when the concept is known, requiring retrieval of a learned visual grounding from text. It remains hard for open ad-hoc concepts, where the grounding m…
EnvFactory: Scaling Tool-Use Agents via Executable Environments Synthesis and Robust RL
Minrui Xu, Zilin Wang, Mengyi DENG +12
Equipping LLMs with tool-use capabilities via Agentic Reinforcement Learning (Agentic RL) is bottlenecked by two challenges: the lack of scalable, robust execution environments and…
DiscoGen: Procedural Generation of Algorithm Discovery Tasks in Machine Learning
Alexander D. Goldie, Zilin Wang, Adrian Hayler +17
Automating the development of machine learning algorithms has the potential to unlock new breakthroughs. However, our ability to improve and evaluate algorithm discovery systems ha…
Smoothness Adaptivity in Constant-Depth Neural Networks: Optimal Rates via Smooth Activations
Yuhao Liu, Zilin Wang, Lei Wu +1
Smooth activation functions are ubiquitous in modern deep learning, yet their theoretical advantages over non-smooth counterparts remain poorly understood. In this work, we study b…
Optimal Learning Rate Schedules under Functional Scaling Laws: Power Decay and Warmup-Stable-Decay
Binghui Li, Zilin Wang, Fengling Chen +3
We study optimal learning rate (LR) schedules under the functional scaling law (FSL) framework (Li et al., 2025), which decomposes training dynamics into signal learning and noise…
Learning to Drive in New Cities Without Human Demonstrations
Zilin Wang, Saeed Rahmani, Daphne Cornelisse +4
While autonomous vehicles have achieved reliable performance within specific operating regions, their deployment to new cities remains costly and slow. A key bottleneck is the need…
Aligning Forest and Trees in Images & Long Captions for Visually Grounded Understanding
Byeongju Woo, Zilin Wang, Byeonghyun Pak +2
Vision-language models such as CLIP often struggle to faithfully understand long, detail-rich captions, relying on dominant scene cues while overlooking fine-grained visual evidenc…
Open Ad-hoc Categorization with Contextualized Feature Learning
Zilin Wang, Sangwoo Mo, Stella X. Yu +2
Adaptive categorization of visual scenes is essential for AI agents to handle changing tasks. Unlike fixed common categories for plants or animals, ad-hoc categories are created dy…
Rethinking the Role of Dynamic Sparse Training for Scalable Deep Reinforcement Learning
Guozheng Ma, Lu Li, Zilin Wang +4
Scaling neural networks has driven breakthrough advances in machine learning, yet this paradigm fails in deep reinforcement learning (DRL), where larger models often degrade perfor…
HyperVLA: Efficient Inference in Vision-Language-Action Models via Hypernetworks
Zheng Xiong, Kang Li, Zilin Wang +3
Built upon language and vision foundation models with strong generalization ability and trained on large-scale robotic data, Vision-Language-Action (VLA) models have recently emerg…
Free-Grained Hierarchical Visual Recognition
Seulki Park, Zilin Wang, Stella X. Yu
Hierarchical image recognition seeks to predict class labels along a semantic taxonomy, from broad categories to specific ones, typically under the tidy assumption that every train…
How Should We Meta-Learn Reinforcement Learning Algorithms?
Alexander David Goldie, Zilin Wang, Jaron Cohen +2
The process of meta-learning algorithms from data, instead of relying on manual design, is growing in popularity as a paradigm for improving the performance of machine learning sys…
Network Sparsity Unlocks the Scaling Potential of Deep Reinforcement Learning
Guozheng Ma, Lu Li, Zilin Wang +3
Effectively scaling up deep reinforcement learning models has proven notoriously difficult due to network pathologies during training, motivating various targeted interventions suc…
Point Cloud Neural Operator for Parametric PDEs on Complex and Variable Geometries
Chenyu Zeng, Yanshu Zhang, Jiayi Zhou +5
Surrogate models are critical for accelerating computationally expensive simulations in science and engineering, particularly for solving parametric partial differential equations…
Dialogue Director: Bridging the Gap in Dialogue Visualization for Multimodal Storytelling
Min Zhang, Zilin Wang, Liyan Chen +2
Recent advances in AI-driven storytelling have enhanced video generation and story visualization. However, translating dialogue-centric scripts into coherent storyboards remains a…
Improving Generalization and Convergence by Enhancing Implicit Regularization
Mingze Wang, Jinbo Wang, Haotian He +6
In this work, we propose an Implicit Regularization Enhancement (IRE) framework to accelerate the discovery of flat solutions in deep learning, thereby improving generalization and…
Explore 3D Dance Generation via Reward Model from Automatically-Ranked Demonstrations
Zilin Wang, Haolin Zhuang, Lu Li +6
This paper presents an Exploratory 3D Dance generation framework, E3D2, designed to address the exploration capability deficiency in existing music-conditioned 3D dance generation…
Distance-rank Aware Sequential Reward Learning for Inverse Reinforcement Learning with Sub-optimal Demonstrations
Lu Li, Yuxin Pan, Ruobing Chen +4
Inverse reinforcement learning (IRL) aims to explicitly infer an underlying reward function based on collected expert demonstrations. Considering that obtaining expert demonstratio…
Normalization Enhances Generalization in Visual Reinforcement Learning
Lu Li, Jiafei Lyu, Guozheng Ma +4
Recent advances in visual reinforcement learning (RL) have led to impressive success in handling complex tasks. However, these methods have demonstrated limited generalization capa…
Learning Better with Less: Effective Augmentation for Sample-Efficient Visual Reinforcement Learning
Guozheng Ma, Linrui Zhang, Haoyu Wang +6
Data augmentation (DA) is a crucial technique for enhancing the sample efficiency of visual reinforcement learning (RL) algorithms. Notably, employing simple observation transforma…
Watch-And-Help: A Challenge for Social Perception and Human-AI Collaboration
Xavier Puig, Tianmin Shu, Shuang Li +5
In this paper, we introduce Watch-And-Help (WAH), a challenge for testing social intelligence in agents. In WAH, an AI agent needs to help a human-like agent perform a complex hous…