Mixture-of-Experts (MoE) large language models reduce computation through sparse activation but still incur substantial storage costs. Mixed-precision post-training quantization (PTQ) reduces these costs by quantizing different model components at different bit widths, but existing methods often struggle to preserve ge...
Jia-Qi Chen, Xiang-Long Yan, Cheng-Zhu Bao et al.· 0 citations
Recently, AutoRegressive Visual Generation (ARVG) models have emerged as a competitive paradigm for high-quality image synthesis. However, their substantial computational and memory demands make them difficult to deploy on resource-constrained edge devices. Post-Training Quantization (PTQ) provides a promising solution...
Xin-Meng Liu, Xiang-Long Yan, Kai-Cheng Yang et al.· 0 citations
Extreme low-bit weight-only quantization is an effective approach to reducing the deployment costs of large language models. Below 3 bits, scalar quantization often suffers severe accuracy loss, while vector quantization (VQ) better preserves accuracy through more expressive joint encoding. However, existing VQ methods...
Xiang-Long Yan, Ze-Nan Wu, Ke-Wei Zhang et al.· 0 citations
It is observed that interaction-induced feature changes correlate with approximation error, while low-frequency interpolation errors are phase-sensitive and show more predictable phase progression, which motivates CAST, a reconstruction-coupled inference framework.
Le-Yang Chen, Jun-Yi Wu, Fan-Qing Kong et al.· 0 citations
Feed-forward 3D vision models such as VGGT have achieved remarkable progress, unifying camera estimation and dense scene reconstruction in a single pass. However, their quadratic global attention makes long image sequences expensive, while existing sparse methods may favor highly attended yet value-redundant regions. T...
Jun-Yi Wu, Fan-Qing Kong, Le-Yang Chen et al.· 0 citations
LT-OPD, a training framework for extreme visual-token reduction, is proposed and it is shown that on-policy learning can substantially recover capabilities lost to extreme visual-token reduction.
Junxian Li, Rui-Xuan Yang, Tian-Ao Zhang et al.· 0 citations
JustQuant is proposed, a simple yet effective framework that moves the complexity of low-bit quantization from deployment-time operators into the training process and introduces Theseus QAD, a quantization-aware distillation method that progressively applies multi-level supervision, analogous to the gradual replacement...
Kai-Cheng Yang, Kai-Sen Yang, Chun-Yu Liu et al.· 0 citations
SPHQuant is a rotation-free spherical weight-only quantization framework for VLMs that isolates outlier magnitude into the radius while keeping directions bounded and statistically regular and designs a hardware-friendly GEMV kernel that keeps the direction codebook small enough for shared-memory lookup and packs radia...
FOCUS is proposed, a post-training quantization framework with end-to-end scale learning for FP4 Optimization via Coupled-Relaxation and Dual-Granularity Scaling, which relaxes the tight coupling between quantization and dequantization scales with a learnable full-precision coefficient, enabling more effective optimiza...
Xiang-Long Yan, Hong Liu, Cheng-Zhu Bao et al.· 1 citation
Visual token pruning reduces the inference cost of multimodal large language models, but a fixed token ratio is poorly matched to text-rich inputs. In OCR-centric tasks, decisive evidence can be a small number, label, or field whose relevance is specified by the question; indiscriminate pruning can erase that evidence...
Zizhong Ding, Junxian Li, Kai Liu 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.
