# Edges Before Embeddings: A Confidence-Aware Blur Gate for Vision-Language Pipelines

arXiv: https://arxiv.org/abs/2606.25838 · HTML: https://mlsanity.com/paper/2606.25838/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Placed in Visual token pruning and efficiency, one of 16 directions in Multimodal LLMs / vision-language models.
- Builds on 9 of its 18 references that are in the map.
- First posted on arXiv 24 Jun 2026.

## Placed in

- Multimodal LLMs / vision-language models / [Visual token pruning and efficiency](https://mlsanity.com/mllm/visual-token-pruning-and-efficiency/)

## Builds on (most cited first, 9 of 9)

- [MUSIQ: Multi-scale Image Quality Transformer](https://mlsanity.com/paper/2108.05997/) (2021)
- [Selective Classification for Deep Neural Networks](https://mlsanity.com/paper/1705.08500/) (2017)
- [Deep Multi-scale Convolutional Neural Network for Dynamic Scene Deblurring](https://mlsanity.com/paper/1612.02177/) (2016)
- [MANIQA: Multi-dimension Attention Network for No-Reference Image Quality Assessment](https://mlsanity.com/paper/2204.08958/) (2022)
- [DeblurGAN: Blind Motion Deblurring Using Conditional Adversarial Networks](https://mlsanity.com/paper/1711.07064/) (2017)
- [Human-Aware Motion Deblurring](https://mlsanity.com/paper/2001.06816/) (2020)
- [DocVLM: Make Your VLM an Efficient Reader](https://mlsanity.com/paper/2412.08746/) (2024)
- [Qianfan-OCR: A Unified End-to-End Model for Document Intelligence](https://mlsanity.com/paper/2603.13398/) (2026)
- [Magika: AI-Powered Content-Type Detection](https://mlsanity.com/paper/2409.13768/) (2024)

## Abstract (arXiv)

Production vision pipelines silently degrade on blurry input, wasting compute on downstream OCR, retrieval, and vision-language model (VLM) calls that cannot recover a usable output. We present MagikaDocumentFromPixel, a lightweight, CPU-friendly image quality gate that classifies a single image as sharp, blurred, or uncertain in roughly 7 ms on a single CPU core. The contributions are (i) a recipe selected from a 46-configuration, 8-sweep empirical search that isolates input resolution as the dominant lever and shows architecture capacity only pays off at >= 384 px; (ii) a confidence-aware routing formalism grounded in classical selective prediction; (iii) the Edge Prior Module (EPM), a Laplacian-magnitude auxiliary input channel that gives the network direct access to the spectral evidence that classical blur heuristics rely on and that lifts test F1 by +1.3 points in a matched-env comparison; and (iv) an observation that the gate is one instance of a recurring design pattern that appears independently in Magika content-type detection, risk-controlled OCR with VLMs, and DocVLM. The final recipe MobileNetV3-Large with the EPM trained at 384x384 on paired GoPro Large frames, evaluated with 5-scale test-time augmentation reaches F1 = 0.9803 (AUC 0.9989) with a 17 MB ONNX artifact, improving over our fixed-scale baseline on the same hardware (F1 = 0.9672) by +1.31 points. We are explicit about limitations: results are on a single motion-blur distribution, numbers are from a single seed, and calibration is qualitative rather than measured.