7 papers
: Twin-Critic Training for Token-Level Thoughts in Reinforcement Mid-Training
Nan Qiao, Yebin Yang, Weinong Wang +10
Reinforcement mid-training lets language models learn internal thoughts from unlabeled text, but efficient token-level credit assignment remains challenging. Existing group-relativ…
Robust Estimation of Sparse Numerical Vectors under Local Differential Privacy
Puning Zhao, Zhikun Zhang, Shaowei Wang +5
Local differential privacy (LDP) protocols are vulnerable to poisoning attacks. Existing research have proposed efficient defense strategies for single-item users. However, in prac…
Executable Agentic Memory for GUI Agent
Zerui Qin, Sheng Yue, Xingyuan Hua +2
Modern GUI agents typically rely on a model-centric and step-wise interaction paradigm, where LLMs must re-interpret the UI and re-decide actions at every screen, which is fragile…
Learning to Explore: Scaling Agentic Reasoning via Exploration-Aware Policy Optimization
Xingyuan Hua, Sheng Yue, Ju Ren
Recent advancements in agentic test-time scaling allow models to gather environmental feedback before committing to final actions. A key limitation of existing methods is that they…
AdamO: A Collapse-Suppressed Optimizer for Offline RL
Nan Qiao, Sheng Yue, Shuning Wang +1
Offline reinforcement learning (RL) can fail spectacularly when bootstrapped temporal-difference (TD) updates amplify their own errors, driving the critic toward extreme and unusab…
FOVA: Offline Federated Reinforcement Learning with Mixed-Quality Data
Nan Qiao, Sheng Yue, Ju Ren +1
Offline Federated Reinforcement Learning (FRL), a marriage of federated learning and offline reinforcement learning, has attracted increasing interest recently. Albeit with some ad…
AugFL: Augmenting Federated Learning with Pretrained Models
Sheng Yue, Zerui Qin, Yongheng Deng +3
Federated Learning (FL) has garnered widespread interest in recent years. However, owing to strict privacy policies or limited storage capacities of training participants such as I…