1. P-Tuning v2: Prompt Tuning Can Be Comparable to Fine-tuning Universally Across Scales and Task (ACL 2022)
2. SwiftSage: A Generative Agent with Fast and Slow Thinking for Complex Interactive Tasks (NeurIPS 2023, spotlight)
3. UI-JEPA: Towards Active Perception of User Intent through Onscreen User Activity (Preprint 2024)
4. CAMPHOR: Collaborative Agents for Multi-input Planning and High-Order Reasoning On Device (Preprint 2024)
5. Pause-Tuning for Long-Context Comprehension: A Lightweight Approach to LLM Attention Recalibration (Preprint 2025)
6. EgoNormia: Benchmarking Physical Social Norm Understanding (Preprint 2025)
Research Experience
Joined Prof. Diyi Yang's lab before studying as an undergraduate at Tsinghua University.
Education
1. MS student, EE department, Stanford, Advisor: Diyi Yang
2. Undergraduate at Electrical Engineering Department, Tsinghua University
Background
Master's student at the Electrical Engineering Department, Stanford University. Broadly interested in natural language processing and artificial intelligence.