Related papers: GRADE: Benchmarking Discipline-Informed Reasoning in Image Editing

GRADE: Benchmarking Discipline-Informed Reasoning in Image Editing

URL: http://arxiv.org/abs/2603.12264v1
Date: Thu, 12 Mar 2026 17:59:52 GMT
Title: GRADE: Benchmarking Discipline-Informed Reasoning in Image Editing
Abstract summary: GRADE is the first benchmark to assess discipline-informed knowledge and reasoning in image editing.<n>GRADE comprises 520 carefully curated samples across 10 academic domains.
Score: 67.29558599215902
License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Abstract: Unified multimodal models target joint understanding, reasoning, and generation, but current image editing benchmarks are largely confined to natural images and shallow commonsense reasoning, offering limited assessment of this capability under structured, domain-specific constraints. In this work, we introduce GRADE, the first benchmark to assess discipline-informed knowledge and reasoning in image editing. GRADE comprises 520 carefully curated samples across 10 academic domains, spanning from natural science to social science. To support rigorous evaluation, we propose a multi-dimensional evaluation protocol that jointly assesses Discipline Reasoning, Visual Consistency, and Logical Readability. Extensive experiments on 20 state-of-the-art open-source and closed-source models reveal substantial limitations in current models under implicit, knowledge-intensive editing settings, leading to large performance gaps. Beyond quantitative scores, we conduct rigorous analyses and ablations to expose model shortcomings and identify the constraints within disciplinary editing. Together, GRADE pinpoints key directions for the future development of unified multimodal models, advancing the research on discipline-informed image editing and reasoning. Our benchmark and evaluation code are publicly released.

Related papers

InEdit-Bench: Benchmarking Intermediate Logical Pathways for Intelligent Image Editing Models [17.680767010203308]
We introduce InEdit-Bench, the first evaluation benchmark dedicated to reasoning over intermediate pathways in image editing.<n>InEdit-Bench comprises meticulously annotated test cases covering four fundamental task categories: state transition, dynamic process, temporal sequence, and scientific simulation.<n>Our comprehensive evaluation of 14 representative image editing models on InEdit-Bench reveals significant and widespread shortcomings in this domain.
arXiv Detail & Related papers (2026-03-04T02:24:43Z)
Human-Aligned MLLM Judges for Fine-Grained Image Editing Evaluation: A Benchmark, Framework, and Analysis [95.89328387635176]
We introduce a fine-grained Multimodal Large Language Model (MLLM)-as-a-Judge framework for image editing.<n>We present a new human-validated benchmark that integrates human judgments, MLLM-based evaluations, model outputs, and traditional metrics.
arXiv Detail & Related papers (2026-02-13T15:34:32Z)
How Well Do Models Follow Visual Instructions? VIBE: A Systematic Benchmark for Visual Instruction-Driven Image Editing [56.60465182650588]
We introduce three-level interaction hierarchy that captures deictic grounding, morphological manipulation, and causal reasoning.<n>We propose a robust LMM-as-a-judge evaluation framework with task-specific metrics to enable scalable and fine-grained assessment.<n>We find that proprietary models exhibit early-stage visual instruction-following capabilities and consistently outperform open-source models.
arXiv Detail & Related papers (2026-02-02T09:24:45Z)
EditThinker: Unlocking Iterative Reasoning for Any Image Editor [72.28251670314451]
We propose a deliberative editing framework to 'think' while they edit.<n>We train a single MLLM, EditThinker, to act as the reasoning engine of this framework.<n>We employ reinforcement learning to align the EditThinker's thinking with its editing, thereby generating more targeted instruction improvements.
arXiv Detail & Related papers (2025-12-05T18:58:09Z)
KRIS-Bench: Benchmarking Next-Level Intelligent Image Editing Models [88.58758610679762]
We introduce KRIS-Bench (Knowledge-based Reasoning in Image-editing Systems Benchmark), a diagnostic benchmark designed to assess models through a cognitively informed lens.<n>We categorize editing tasks across three foundational knowledge types: Factual, Conceptual, and Procedural.<n>To support fine-grained evaluation, we propose a protocol that incorporates a novel Knowledge Plausibility metric, enhanced by knowledge hints and calibrated through human studies.
arXiv Detail & Related papers (2025-05-22T14:08:59Z)
Envisioning Beyond the Pixels: Benchmarking Reasoning-Informed Visual Editing [84.16442052968615]
We introduce RISEBench, the first benchmark for evaluating Reasoning-Informed viSual Editing (RISE)<n>RISEBench focuses on four key reasoning categories: Temporal, Causal, Spatial, and Logical Reasoning.<n>We conduct experiments evaluating nine prominent visual editing models, comprising both open-source and proprietary models.
arXiv Detail & Related papers (2025-04-03T17:59:56Z)
PixLens: A Novel Framework for Disentangled Evaluation in Diffusion-Based Image Editing with Object Detection + SAM [17.89238060470998]
evaluating diffusion-based image-editing models is a crucial task in the field of Generative AI. Our benchmark, PixLens, provides a comprehensive evaluation of both edit quality and latent representation disentanglement.
arXiv Detail & Related papers (2024-10-08T06:05:15Z)
Counterfactual Edits for Generative Evaluation [0.0]
We propose a framework for the evaluation and explanation of synthesized results based on concepts instead of pixels. Our framework exploits knowledge-based counterfactual edits that underline which objects or attributes should be inserted, removed, or replaced from generated images. Global explanations produced by accumulating local edits can also reveal what concepts a model cannot generate in total.
arXiv Detail & Related papers (2023-03-02T20:10:18Z)

This list is automatically generated from the titles and abstracts of the papers in this site.