SPHQuant: Efficient extreme low bit weight quantization for Vision-Language Models
cs.CV
Submitted: 2026-09-21
Updated: 2026-09-21
Code: https://github.com/Pushazf/SPHQuant
Terminology
Sources
- Qwen3-VL Technical Report
- Are We on the Right Way for Evaluating Large Vision-Language Models?
- QuantDemoire: Quantization with Outlier Aware for Image Demoir'eing
- MobileVLM : A Fast, Strong and Open Vision Language Assistant for Mobile Devices
- MobileVLM V2: Faster and Stronger Baseline for Vision Language Model
- GPTQ: Accurate Post-Training Quantization for Generative Pre-trained Transformers
- Efficient Multimodal Large Language Models: A Survey
- Adam: A Method for Stochastic Optimization
- MBQ: Modality-Balanced Quantization for Large Vision-Language Models
- Microsoft COCO: Common Objects in Context
- Learn to Explain: Multimodal Reasoning via Thought Chains for Science Question Answering
- VEQ: Modality-Adaptive Quantization for MoE Vision-Language Models
- Towards VQA Models That Can Read
- QuIP#: Even Better LLM Quantization with Hadamard Incoherence and Lattice Codebooks
- Pyramid Vector Quantization for LLMs
- InternVL3.5: Advancing Open-Source Multimodal Models in Versatility, Reasoning, and Efficiency
- LSGQuant: Layer-Sensitivity Guided Quantization for One-Step Diffusion Real-World Video Super-Resolution
- MMMU: A Massive Multi-discipline Multimodal Understanding and Reasoning Benchmark for Expert AGI
- MMMU-Pro: A More Robust Multi-discipline Multimodal Understanding Benchmark
- PCDVQ: Enhancing Vector Quantization for Large Language Models via Polar Coordinate Decoupling
Related papers
- Loss Knows Best: Detecting Annotation Errors in Videos via Loss Trajectories
- AnchorWeave: World-Consistent Video Generation with Retrieved Local Spatial Memories
- Benchmarking the Robustness of Foundation Models for Mammography under Domain Shift
- MambaX-Net: Dual-Input Mamba-Enhanced Cross-Attention Network for Longitudinal MRI Segmentation
- TeleOCR: Navigating Document Parsing Across Digital and Camera-Captured Documents
- A Survey on Efficient Vision-Language-Action Models