Profiles2024-2026
Zhongxiang Dai
The Chinese University of Hong Kong, Shenzhen
15Papers
6Last author
22Unique co-authors
0First author
0Solo
First and last author come from byline position, which some venues assign alphabetically.
Recent papers15
- ActiveDPO: Active Direct Preference Optimization for Sample-Efficient AlignmentAuthor 3 of 7ICLR 2026 · 12 citations
- Large Language Model-Enhanced Multi-Armed BanditsAuthor 6 of 6ACL 2026 · 6 citations
- T-POP: Test-Time Personalization with Online Preference FeedbackAuthor 10 of 10ICML 2026 · 4 citations
- MASPOB: Bandit-Based Prompt Optimization for Multi-Agent Systems with Graph Neural NetworksAuthor 8 of 8ICML 2026 · 3 citations
- ALSO: Adversarial Online Strategy Optimization for Social AgentsAuthor 5 of 6ICML 2026
- Federated Linear Dueling BanditsAuthor 6 of 6AAAI 2026
- UniScale: Adaptive Unified Inference Scaling via Online Joint Optimization of Model Routing and Test-Time ScalingAuthor 7 of 9ICML 2026
- Adaptive Batch-Wise Sample Scheduling for Direct Preference OptimizationAuthor 5 of 7NeurIPS 2025 · 14 citations
- Convergence Rates of Constrained Expected ImprovementAuthor 4 of 6NeurIPS 2025 · 3 citations
- Neural Dueling Bandits: Preference-Based Optimization with Human FeedbackAuthor 2 of 6ICLR 2025
- Online Clustering of Dueling BanditsAuthor 7 of 7ICML 2025
- Refining Adaptive Zeroth-Order Optimization at EaseAuthor 4 of 4ICML 2025
- WASA: WAtermark-based Source Attribution for Large Language Model-Generated DataAuthor 4 of 8ACL Findings 2025
- Prompt Optimization with EASE? Efficient Ordering-aware Automated Selection of ExemplarsAuthor 3 of 9NeurIPS 2024 · 44 citations
- Localized Zeroth-Order Prompt OptimizationAuthor 6 of 9NeurIPS 2024 · 30 citations
