Vision Transformers (ViTs) achieve state-of-the-art performance across various vision tasks but pose significant deployment challenges due to their high computational complexity and memory footprint. Post-training quantization (PTQ) offers an efficient, retraining-free compression strategy; however, existing solutions...
Ning Yang, Chen-Yang Guan, Yuang Peng et al.· ACM Transactions on Architec...· 0 citations
This paper proposes VQVLA, an algorithm-hardware co-design framework that accelerates VLA inference by exploiting weight similarity and execution dynamics, and proposes a merged-centroid vectorized GEMM paradigm that operates on the codebook-index representation.
Comprehensive evaluations on OpenVLA and RDT across LIBERO and ManiSkill benchmarks show that SpecVLA reduces end-to-end latency significantly while preserving task success rate, and achieves real-time robotic manipulation with both high efficiency and reliability.
Chunyu Qi, Zhuoran Song, Jian Weng et al.· 0 citations
PCoMoE is presented, a path-compositional execution framework that shifts MoE inference from coarse-grained expert selection to fine-grained path composition and achieves up to a 1.31x end-to-end inference speedup while enhancing model accuracy by 10%.
Zi-Yan Gan, Fang-Xin Liu, Chenyang Guan et al.· 0 citations
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.