Qi Lei

6

Papers

4

Total Citations

Papers (6)

Beyond Interpretability: The Gains of Feature Monosemanticity on Model Robustness

Mono3DVLT: Monocular-Video-Based 3D Visual Language Tracking

Beyond Losses Reweighting: Empowering Multi-Task Learning via the Generalization Perspective

Performative Risk Control: Calibrating Models for Reliable Deployment under Performativity

Controllable Prompt Tuning For Balancing Group Distributional Robustness

An Information-Theoretic Analysis of In-Context Learning