LASER introduces curvature-weighted SVD from second-order loss approximation and loss-aware rank allocation to compress VLMs, reporting over 2.3x decoding speedup under low-precision settings.
Dfrot: Achieving outlier-free and massive activation-free for rotated llms with refined rotation.arXiv preprint arXiv:2412.00648
3 Pith papers cite this work. Polarity classification is still indexing.
3
Pith papers citing it
years
2026 3verdicts
UNVERDICTED 3representative citing papers
RUQuant uses block-wise composite orthogonal matrices from Householder reflections and Givens rotations plus a fine-tuned global reflection to achieve 99.8% full-precision accuracy at W6A6 and 97% at W4A4 for 13B LLMs in about one minute.
WSVD delivers over 1.8x faster VLM decoding via weighted low-rank approximation at fine granularity plus quantization, without accuracy loss.
citing papers explorer
-
LASER: Loss-Aware Singular-value Decomposition and Rank Allocation for Efficient Low-Precision Vision-Language Models
LASER introduces curvature-weighted SVD from second-order loss approximation and loss-aware rank allocation to compress VLMs, reporting over 2.3x decoding speedup under low-precision settings.
-
RUQuant: Towards Refining Uniform Quantization for Large Language Models
RUQuant uses block-wise composite orthogonal matrices from Householder reflections and Givens rotations plus a fine-tuned global reflection to achieve 99.8% full-precision accuracy at W6A6 and 97% at W4A4 for 13B LLMs in about one minute.
-
WSVD: Weighted Low-Rank Approximation for Fast and Efficient Execution of Low-Precision Vision-Language Models
WSVD delivers over 1.8x faster VLM decoding via weighted low-rank approximation at fine granularity plus quantization, without accuracy loss.