Pingcheng Dong

发表论文 22 篇 · 总被引 199 次 · h-index 6

代表论文

LLM-FP4: 4-Bit Floating-Point Quantized Transformers (2023 · Conference on Empirical Methods in Natural Language Processing · 被引 130)
Quantization Variation: A New Perspective on Training Transformers with Low-Bit Precision (2023 · Trans. Mach. Learn. Res. · 被引 26)
A 28nm 0.22μJ/Token Memory-Compute-Intensity-Aware CNN-Transformer Accelerator with Hybrid-Attention-Based Layer-Fusion and Cascaded Pruning for Semantic-Segmentation (2025 · IEEE International Solid-State Circuits Conference · 被引 24)
Genetic Quantization-Aware Approximation for Non-Linear Operations in Transformers (2024 · Design Automation Conference · 被引 19)
Cyclic Contrastive Knowledge Transfer for Open-Vocabulary Object Detection (2025 · International Conference on Learning Representations · 被引 12)
Memory Efficient Transformer Adapter for Dense Predictions (2025 · International Conference on Learning Representations · 被引 7)