# Towards Characterizing Scientific Image Utility and Upgradability

arXiv: https://arxiv.org/abs/2606.03401 · HTML: https://mlsanity.com/paper/2606.03401/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Builds on 26 of its 39 references that are in the map.
- First posted on arXiv 2 Jun 2026.

## Builds on (most cited first, 26 of 26)

- [High-Resolution Image Synthesis with Latent Diffusion Models](https://mlsanity.com/paper/2112.10752/) (2021)
- [InstructPix2Pix: Learning to Follow Image Editing Instructions](https://mlsanity.com/paper/2211.09800/) (2022)
- [Learn to Explain: Multimodal Reasoning via Thought Chains for Science Question Answering](https://mlsanity.com/paper/2209.09513/) (2022)
- [InternVL3.5: Advancing Open-Source Multimodal Models in Versatility, Reasoning, and Efficiency](https://mlsanity.com/paper/2508.18265/) (2025)
- [Galactica: A Large Language Model for Science](https://mlsanity.com/paper/2211.09085/) (2022)
- [Q-Align: Teaching LMMs for Visual Scoring via Discrete Text-Defined Levels](https://mlsanity.com/paper/2312.17090/) (2023)
- [Kimi K2.5: Visual Agentic Intelligence](https://mlsanity.com/paper/2602.02276/) (2026)
- [Step1X-Edit: A Practical Framework for General Image Editing](https://mlsanity.com/paper/2504.17761/) (2025)
- [SciBench: Evaluating College-Level Scientific Problem-Solving Abilities of Large Language Models](https://mlsanity.com/paper/2307.10635/) (2023)
- [Deep Neural Networks for No-Reference and Full-Reference Image Quality Assessment](https://mlsanity.com/paper/1612.01697/) (2016)
- [KOSMOS-2.5: A Multimodal Literate Model](https://mlsanity.com/paper/2309.11419/) (2023)
- [InstructMol: Multi-Modal Integration for Building a Versatile and Reliable Molecular Assistant in Drug Discovery](https://mlsanity.com/paper/2311.16208/) (2023)
- [ChemVLM: Exploring the Power of Multimodal Large Language Models in Chemistry Area](https://mlsanity.com/paper/2408.07246/) (2024)
- [DeepScientist: Advancing Frontier-Pushing Scientific Findings Progressively](https://mlsanity.com/paper/2509.26603/) (2025)
- [Learning Generalized Spatial-Temporal Deep Feature Representation for No-Reference Video Quality Assessment](https://mlsanity.com/paper/2012.13936/) (2020)
- [A-Bench: Are LMMs Masters at Evaluating AI-generated Images?](https://mlsanity.com/paper/2406.03070/) (2024)
- [Q-Eval-100K: Evaluating Visual Quality and Alignment Level for Text-to-Vision Content](https://mlsanity.com/paper/2503.02357/) (2025)
- [LMM4Edit: Benchmarking and Evaluating Multimodal Image Editing with LMMs](https://mlsanity.com/paper/2507.16193/) (2025)
- [GenExam: A Multidisciplinary Text-to-Image Exam](https://mlsanity.com/paper/2509.14232/) (2025)
- [BMMR: A Large-Scale Bilingual Multimodal Multi-Discipline Reasoning Dataset](https://mlsanity.com/paper/2507.03483/) (2025)
- [I2I-Bench: A Comprehensive Benchmark Suite for Image-to-Image Editing Models](https://mlsanity.com/paper/2512.04660/) (2025)
- [GRADE: Benchmarking Discipline-Informed Reasoning in Image Editing](https://mlsanity.com/paper/2603.12264/) (2026)
- [OmniScience: A Large-scale Multi-modal Dataset for Scientific Image Understanding](https://mlsanity.com/paper/2602.13758/) (2026)
- [SIQA: Toward Reliable Scientific Image Quality Assessment](https://mlsanity.com/paper/2603.06700/) (2026)
- [Plug In, Grade Right: Psychology-Inspired AGIQA](https://mlsanity.com/paper/2512.22780/) (2025)
- [From Global to Granular: Revealing IQA Model Performance via Correlation Surface](https://mlsanity.com/paper/2601.21738/) (2026)

## Built on by (most cited first, 1 of 1)

- [SciFigAlign: Scoring Scientific Figures by Fine-tuned Alignment of Visuals with Manuscript Evidence](https://mlsanity.com/paper/2607.27066/) (2026)

## Abstract (arXiv)

Scientific images function as critical evidence in research communication, yet their integrity faces unprecedented threats from AI-generated content that introduces subtle but consequential errors. Existing evaluation paradigms prove inadequate: perceptual quality metrics poorly correlate with scientific validity, while language models lack domain-specific verification capabilities. To address this gap, we propose the Scientific Image Utility and Upgradability Assessment (SIU2A) framework, which introduces two complementary dimensions for scientific image evaluation. Utility encompasses textiterror detection (identifying scientific inaccuracies) and textitcorrection feasibility (assessing whether errors can be reliably repaired). Upgradability measures the quality of correction. We categorize scientific image corruption into four fundamental types: Detail Distortion, Incompleteness, False Content, and Entity Confusion. Based on this taxonomy, we construct SIU2A-Benchmark, a dataset with expert annotations for error identification and repair. The framework implements a two-stage evaluation protocol: the textitUtility stage evaluates error detection capability and repair instruction generation, while the textitUpgradability stage assesses whether corrections faithfully restore scientific validity without compromising existing accurate information. Experiments reveal that current multimodal systems exhibit significant limitations in both scientific error assessment and faithful correction, exposing a fundamental gap between visual perception and scientific usability.