Shihao Ji

6

Papers

2

Total Citations

Papers (6)

Uni-LoRA: One Vector is All You Need

NeurIPS 2025arXiv

Unsqueeze [CLS] Bottleneck to Learn Rich Representations

Chitransformer: Towards Reliable Stereo From Cues

Towards Bridging the Performance Gaps of Joint Energy-Based Models

JEM++: Improved Techniques for Training JEM

FLSL: Feature-level Self-supervised Learning