FIMA-Q
FIMA-Q: Post-Training Quantization for Vision Transformers by Fisher Information Matrix ApproximationLLM quantization · first seen Jun 13, 2025
superseded — cited as a baseline and beaten by newer methods
2 papers critique it · 1 beat it on benchmarks
What papers say
Verbatim critique sentences, each from a paper that cites FIMA-Q as a baseline.
“FIMA-Q wu2025fima proposes an efficient Fisher Information Matrix approximation to guide quantization via block-wise reconstruction loss, pushing ViT PTQ toward W3A3 and W4A4 with notable improvement. However, the block-level independence neglects inter-block compensation, leaving further potential unexploited.”
— Joint Post-Training Quantization of Vision Transformers with Learned Prompt-Guided Data Generation“These strategies are largely layer or block-centric and optimize average fidelity. In COD at W4A4, the dominant failure is token-local: token-wise activation heterogeneity lets background tokens dominate the range and increase the zero-bin mass, which layer-wise fitting or reconstruction does not explicitly bound.”
— When W4A4 Breaks Camouflaged Object Detection: Token-Group Dual-Constraint Activation Quantization
Beaten on benchmarks
Head-to-head results where a newer method reports beating FIMA-Q. Values are copied from the source paper's tables — verify against the cited paper.
- Joint Post-Training Quantization of Vision Transformers with Learned Prompt-Guided Data Generation
Joint Post-Training Quantization of Vision Transformers with Learned Prompt-Guided Data Generation beats FIMA-Q · Top-1 accuracy [ViT-S, W1.58A8]
68.45 vs 4.84
- Joint Post-Training Quantization of Vision Transformers with Learned Prompt-Guided Data Generation
Joint Post-Training Quantization of Vision Transformers with Learned Prompt-Guided Data Generation beats FIMA-Q · Top-1 accuracy [ViT-S, W3A3]
71.89 vs 64.09
- Joint Post-Training Quantization of Vision Transformers with Learned Prompt-Guided Data Generation
Joint Post-Training Quantization of Vision Transformers with Learned Prompt-Guided Data Generation beats FIMA-Q · Top-1 accuracy [ViT-S, W4A4]
78.35 vs 76.68
- Joint Post-Training Quantization of Vision Transformers with Learned Prompt-Guided Data Generation
Joint Post-Training Quantization of Vision Transformers with Learned Prompt-Guided Data Generation beats FIMA-Q · Top-1 accuracy [ViT-S, W6A6]
80.84 vs 80.64
- Joint Post-Training Quantization of Vision Transformers with Learned Prompt-Guided Data Generation
Joint Post-Training Quantization of Vision Transformers with Learned Prompt-Guided Data Generation beats FIMA-Q · Top-1 accuracy [ViT-B, W1.58A8]
78.51 vs 45.55
- Joint Post-Training Quantization of Vision Transformers with Learned Prompt-Guided Data Generation
Joint Post-Training Quantization of Vision Transformers with Learned Prompt-Guided Data Generation beats FIMA-Q · Top-1 accuracy [ViT-B, W3A3]
79.63 vs 77.63
- Joint Post-Training Quantization of Vision Transformers with Learned Prompt-Guided Data Generation
Joint Post-Training Quantization of Vision Transformers with Learned Prompt-Guided Data Generation beats FIMA-Q · Top-1 accuracy [ViT-B, W4A4]
83.47 vs 83.04
- Joint Post-Training Quantization of Vision Transformers with Learned Prompt-Guided Data Generation
Joint Post-Training Quantization of Vision Transformers with Learned Prompt-Guided Data Generation beats FIMA-Q · Top-1 accuracy [DeiT-S, W1.58A8]
70.13 vs 33.93
- Joint Post-Training Quantization of Vision Transformers with Learned Prompt-Guided Data Generation
Joint Post-Training Quantization of Vision Transformers with Learned Prompt-Guided Data Generation beats FIMA-Q · Top-1 accuracy [DeiT-S, W3A3]
71.55 vs 69.13
- Joint Post-Training Quantization of Vision Transformers with Learned Prompt-Guided Data Generation
Joint Post-Training Quantization of Vision Transformers with Learned Prompt-Guided Data Generation beats FIMA-Q · Top-1 accuracy [DeiT-S, W4A4]
77.25 vs 76.87
- Joint Post-Training Quantization of Vision Transformers with Learned Prompt-Guided Data Generation
Joint Post-Training Quantization of Vision Transformers with Learned Prompt-Guided Data Generation beats FIMA-Q · Top-1 accuracy [Swin-S, W1.58A8]
76.23 vs 38.13
- Joint Post-Training Quantization of Vision Transformers with Learned Prompt-Guided Data Generation
Joint Post-Training Quantization of Vision Transformers with Learned Prompt-Guided Data Generation beats FIMA-Q · Top-1 accuracy [Swin-S, W3A3]
78.41 vs 77.26
Newer alternatives
Recent methods in the same sub-problem, not yet superseded in the knowledge base.
- COD-TDQWhen W4A4 Breaks Camouflaged Object Detection: Token-Group Dual-Constraint Activation QuantizationApr 18, 2026
- Joint Post-Training Quantization of Vision Transformers with Learned Prompt-Guided Data GenerationJoint Post-Training Quantization of Vision Transformers with Learned Prompt-Guided Data GenerationFeb 21, 2026