TileQuant: Tile as the Precision-Allocation Unit for LLM Mixed-Precision Post-Training Quantization
cs.LG, cs.AI
Submitted: 2026-11-04
Updated: 2026-11-04
Code: https://github.com/HanzhiZhang-Ulrica/TileQuant
Terminology
Sources
- Quantization Error Propagation: Revisiting Layer-Wise Post-Training Quantization
- DiscQuant: A Quantization Method for Neural Networks Inspired by Discrepancy Theory
- Channel-Wise Mixed-Precision Quantization for Large Language Models
- Think you have Solved Question Answering? Try ARC, the AI2 Reasoning Challenge
- SpQR: A Sparse-Quantized Representation for Near-Lossless LLM Weight Compression
- GPTQ: Accurate Post-Training Quantization for Generative Pre-trained Transformers
- Measuring Massive Multitask Language Understanding
- SliM-LLM: Salience-Driven Mixed-Precision Quantization for Large Language Models
- BoA: Attention-aware Post-training Quantization without Backpropagation
- SqueezeLLM: Dense-and-Sparse Quantization
- Is (Selective) Round-To-Nearest Quantization All You Need?
- BRECQ: Pushing the Limit of Post-Training Quantization by Block Reconstruction
- TesseraQ: Ultra Low-Bit LLM Post-Training Quantization with Block Reconstruction
- QLLM: Accurate and Efficient Low-Bitwidth Quantization for Large Language Models
- SpinQuant: LLM quantization with learned rotations
- Pointer Sentinel Mixture Models
- Mitigating the Impact of Outlier Channels for Language Model Quantization with Activation Regularization
- SmoothQuant+: Accurate and Efficient 4-bit Post-Training WeightQuantization for LLM
- ResQ: Mixed-Precision Quantization of Large Language Models with Low-Rank Residuals
- PB-LLM: Partially Binarized Large Language Models
Related papers
- A Physics-Regulated Neural Framework for Learning 3D Grain Growth Dynamics
- FlexMoE: Learning One-for-All Multi-Budget Expert Pruning for Flexible MoE Deployment
- OptiMind: Improving Small Language Models for Optimization Modeling through Formulation Guidance and Multi-Turn Learning
- Staying in the Sweet Spot: Boost Reasoning Evolution via Responsive Rollout Difficulty Manipulation
- Examining the Impact of Data and Training Methods on Emergent and Subliminal Misalignment
- Rethinking Black-Box On-Policy Distillation: A Rubric-based Analysis