11 citations · 27 across the 22 of their papers we have counts for
8 papers · 1 filter
On the Depth Transfer of Neural Optimizers via L-Smoothness
Maximilian Wong, Chenyu Xu, Jiajin Li +1
Scaling deep learning models efficiently requires predictable performance transfer across compute regimes. Establishing reliable hyperparameter transfer allows optimal training dyn…
When PINNs Go Wrong: Pseudo-Time Stepping Against Spurious Solutions
Sifan Wang, Shawn Koohy, Yiping Lu +1
Physics-informed neural networks (PINNs) provide a promising machine learning framework for solving partial differential equations, but their training often breaks down on challeng…
RAGEN-2: Reasoning Collapse in Agentic RL
Zihan Wang, Chi Gui, Xing Jin +13
RL training of multi-turn LLM agents is inherently unstable, and reasoning quality directly determines task performance. Entropy is widely used to track reasoning stability. Howeve…
On the Width Scaling of Neural Optimizers Under Matrix Operator Norms I: Row/Column Normalization and Hyperparameter Transfer
Ruihan Xu, Jiajin Li, Yiping Lu
A central question in modern deep learning is how to design optimizers whose behavior remains stable as the network width increases. We address this question by interpreting se…
A Simple "Try Again" Can Elicit Multi-Turn LLM Reasoning
Licheng Liu, Zihan Wang, Linjie Li +5
Multi-turn problem solving is critical yet challenging for Large Reasoning Models (LRMs) to reflect on their reasoning and revise from feedback. Existing Reinforcement Learning (RL…
RAGEN: Understanding Self-Evolution in LLM Agents via Multi-Turn Reinforcement Learning
Zihan Wang, Kangrui Wang, Qineng Wang +15
Training large language models (LLMs) as interactive agents presents unique challenges including long-horizon decision making and interacting with stochastic environment feedback.…
Generalization of Scaled Deep ResNets in the Mean-Field Regime
Yihang Chen, Fanghui Liu, Yiping Lu +2
Despite the widespread empirical success of ResNet, the generalization properties of deep ResNet are rarely explored beyond the lazy training regime. In this work, we investigate \…
An Unconstrained Layer-Peeled Perspective on Neural Collapse
Wenlong Ji, Yiping Lu, Yiliang Zhang +2
Neural collapse is a highly symmetric geometric pattern of neural networks that emerges during the terminal phase of training, with profound implications on the generalization perf…