Haoyuan Sun
10 papers · 2019–2026 · 8 venues · across top CS/AI venues
Authors are matched by name, so researchers who share a name ("Yang Liu", "Wei Wang") appear as a single profile and their paper counts, venues and achievements are combined. How identity is handled.
Achievements
Jump to papers ↓+5 more ↓ Show less ↑
🗺️ Taxonomy Completionist (25) 🌍 Conference Polyglot · Silver 🏃 Academic Marathon · Sprint 🐣 Hot Topic Early Bird 🌉 Interdisciplinary Bridge · Silver
🧬
Topic Evolution
🏆
Keyword Champion
🌈
Renaissance Researcher
(5)
🗃️
Keyword Collector
(63)
💎
Century Club · Bronze
Top co-authors
Research topics
Machine Learning
(8)
Deep Learning
(6)
Mathematics & Optimization
(6)
Optimization
(6)
Optimization & Theory
(6)
Learning Types · Deep Learning
(3)
Learning Types · Machine Learning
(3)
Artificial Intelligence
(2)
Robotics
(2)
Core AI
(2)
Systems
(2)
Computer Vision
(1)
Reinforcement Learning
(1)
Architectures
(1)
Core Methods
(1)
Keywords
mirror descent
(3)
gradient descent
(3)
overparameterized model
(2)
maximum-margin solution
(2)
divergence minimization
(1)
curriculum learning
(1)
regret analysis
(1)
in-context learning
(1)
representation learning
(1)
multimodal learning
(1)
text-to-image generation
(1)
neural network optimization
(1)
dynamic regret
(1)
reinforcement learning from human feedback
(1)
online convex optimization
(1)
preference alignment
(1)
bregman divergence
(1)
kernel regression
(1)
maximum margin
(1)
direct preference optimization
(1)
Papers
CoLoGen: Progressive Learning of Concept-Localization Duality for Unified Image Generation
CVPR 2026
On the Role of Transformer Feed-Forward Layers in Nonlinear In-Context Learning
ALT 2026
Entropy-based Activation Function Optimization: A Method on Searching Better Activation Functions
ICLR 2025
FloorPlan-LLaMa: Aligning Architects’ Feedback and Domain Knowledge in Architectural Floor Plan Generation
ACL 2025
Generalizing Alignment Paradigm of Text-to-Image Generation with Preferences Through f-Divergence Minimization
AAAI 2025
Meta-Learning for Adaptive Control with Automated Mirror Descent
L4DC 2025
A Unified Approach to Controlling Implicit Regularization via Mirror Descent
JMLR 2023
Mirror Descent Maximizes Generalized Margin and Can Be Implemented Efficiently
NeurIPS 2022
Perturbation-based Regret Analysis of Predictive Control in Linear Time Varying Systems
NeurIPS 2021
Beyond Online Balanced Descent: An Optimal Algorithm for Smoothed Online Optimization
NeurIPS 2019