Xunliang Cai
33 papers · 203 citations
Most-cited papers
- Speculative Decoding Via Early-exiting For Faster LLM Inference With Thompson Sampling Control Mechanism2024 · 35 citations
- Large Language Models Meet Open-world Intent Discovery And Recognition: An Evaluation Of Chatgpt2023 · 27 citations
- What Makes Quantization For Large Language Models Hard? An Empirical Study From The Lens Of Perturbation2024 · 22 citations
- Preference Curriculum: Llms Should Always Be Pretrained On Their Preferred Data2025 · 14 citations
- Can RL Improve Generalization Of LLM Agents? An Empirical Study2026
- Exploring Reasoning Reward Model For Agents2026
- Onethinker: All-in-one Reasoning Model For Image And Video2026
- DORA: A Scalable Asynchronous Reinforcement Learning System For Language Model Training2026
- Amemgym: Interactive Memory Benchmarking For Assistants In Long-horizon Conversations2026
Topics