跳到正文
HuggingFace Daily Papers(社區熱門論文)·· 2026-10-01AI 評分43

VIEScore2: Unified Image Evaluation with Spatially Grounded Explanations

VIEScore2: Unified Image Evaluation with Spatially Grounded Explanations

AI 導讀

VIEScore2 是用於圖像生成與編輯任務的統一評估器,可選擇使用條件圖像;它以單一模型推理同步預測品質評分及缺陷位置。在主要測試套件中,VIEScore2 的整體評分 SRCC 為 0.601,高於 Gemini-3-Flash 的 0.491;缺陷定位則在六個基準中的三個勝過通用視覺語言模型及專用空間評估器,並在五個基準排名前三。

正文

Published on Oct 1

·

Submitted by

Max Ku

on Oct 8

Authors:

,

,

,

,

,

,

Abstract

Existing synthetic image evaluators typically provide only a scalar quality score and do not identify the image regions that support it. We introduce VIEScore2, a unified evaluator for image generation and editing tasks with optional conditioning images. VIEScore2 represents an image as an N x N grid and jointly predicts quality scores and defect locations in a single model pass. Its text-native grid representation provides a common interface for heterogeneous spatial supervision and enables directly verifiable post-training objectives. We train on 38K examples spanning score-only, localization-only, and joint supervision across generation and editing tasks. Starting from supervised fine-tuning, we further apply GRPO to improve defect localization using rewards that combine cell-level Dice overlap, score accuracy, and output-format validity. A parameter-free parser converts the structured predictions into readable explanations. On the primary suite, VIEScore2 achieves an overall-score SRCC of 0.601, compared with 0.491 for Gemini-3-Flash, the strongest zero-shot general-purpose VLM baseline under matched inputs. For defect localization, VIEScore2 outperforms both general-purpose VLMs and specialized spatial evaluators on three of six benchmarks in per-image grid IoU and ranks among the top three on five, including datasets beyond its training sources.

View arXiv page View PDF Project page GitHub Add to collection

Get this paper in your agent:

hf papers read 2610.00994

Don't have the latest CLI?

curl -LsSf https://hf.co/cli/install.sh | bash

Models citing this paper 1

Allenda/VIEScore2

Image-Text-to-Text •

9B •

Updated 33 minutes ago

•

39

Datasets citing this paper 0

No dataset linking this paper

Cite arxiv.org/abs/2610.00994 in a dataset README.md to link it from this page.

Spaces citing this paper 0

No Space linking this paper

Cite arxiv.org/abs/2610.00994 in a Space README.md to link it from this page.

Collections including this paper 0

No Collection including this paper

Add this paper to a collection to link it from this page.

來源:HuggingFace Daily Papers(社區熱門論文) · huggingface.co