Skip to content
AI News HubLIVE
Original source2 min read

Beyond Blind Compliance: Benchmarking Task Verification in OCR Reasoning

Summary

Multimodal large language models (MLLMs) excel on OCR document understanding and text-rich visual reasoning benchmarks, but existing evaluations often assume all tasks are valid and answerable. In reality, questions may rely on illegible text, occluded evidence, nonexistent visual targets, contradictory premises, or missing variables. This paper introduces VeriOCRBench, a human-verified benchmark with 1,800 samples, including 1,600 trap-embedded invalid tasks and 200 trap-free controls, to evaluate task verification, root-cause diagnosis, and over-refusal. Evaluation of 15 leading MLLMs reveals persistent blind compliance, diagnosis failures, and prompt-induced over-refusal, highlighting a critical reliability gap in current OCR reasoning systems.

SourcearXiv Computer VisionAuthor: Yue Zhou, Yuan Wu, Yi Chang
Beyond Blind Compliance: Benchmarking Task Verification in OCR Reasoning
Report an error

The correction channel is not available yet. You can copy the article reference below for later.

Correction instructions
Read article

[Submitted on 31 Aug 2026]

Title:Beyond Blind Compliance: Benchmarking Task Verification in OCR Reasoning

View a PDF of the paper titled Beyond Blind Compliance: Benchmarking Task Verification in OCR Reasoning, by Yue Zhou and 2 other authors

View PDF HTML (experimental)

Abstract:Multimodal Large Language Models (MLLMs) have achieved strong performance on OCR-centric document understanding and text-rich visual reasoning benchmarks. Yet existing evaluations largely assume that every task is valid and answerable. In real-world OCR scenarios, this assumption often fails: questions may rely on illegible text, occluded evidence, nonexistent visual targets, contradictory premises, or missing variables. We study this reliability gap as OCR-grounded Task Verification: before answering, a model should determine whether the Image Premise (IP), Textual Premise (TP), and Question (Q) jointly define an executable task.

We introduce VeriOCRBench, a 1,800-sample human-verified benchmark built from source images drawn from 8 OCR-related datasets and spanning 8 real-world image domains, with controlled, image-grounded diagnostic tasks.

It contains 1,600 trap-injected invalid tasks across 8 trap types and four verification dimensions---Visual, Contextual, Factual, and Logical---plus 200 trap-free controls for measuring over-refusal. Built with a Visual Atomic Fact (VAF)-anchored pipeline and full human auditing, VeriOCRBench enables decoupled evaluation of task verification, root-cause diagnosis, and over-refusal. Evaluating 15 leading MLLMs reveals persistent blind compliance, diagnosis failures, and prompt-induced over-refusal, exposing a critical reliability gap in current OCR reasoning systems.

The code is available at: this https URL.

Subjects:

Computer Vision and Pattern Recognition (cs.CV)

Cite as: arXiv:2609.00232 [cs.CV]

(or arXiv:2609.00232v1 [cs.CV] for this version)

https://doi.org/10.48550/arXiv.2609.00232

arXiv-issued DOI via DataCite (pending registration)

Submission history

From: Yue Zhou [view email] [v1] Mon, 31 Aug 2026 18:39:46 UTC (6,095 KB)

Full-text links:

Access Paper:

View a PDF of the paper titled Beyond Blind Compliance: Benchmarking Task Verification in OCR Reasoning, by Yue Zhou and 2 other authors

View PDF

HTML (experimental)

TeX Source

view license

Current browse context:

cs.CV

new | recent | 2026-09

Change to browse by:

cs

References & Citations

NASA ADS

Google Scholar

Semantic Scholar

Loading...

Data provided by:

Bibliographic Tools

Bibliographic and Citation Tools

Bibliographic Explorer Toggle

Bibliographic Explorer (What is the Explorer?)

Connected Papers Toggle

Connected Papers (What is Connected Papers?)

Litmaps Toggle

Litmaps (What is Litmaps?)

scite.ai Toggle

scite Smart Citations (What are Smart Citations?)

Code, Data, Media

Code, Data and Media Associated with this Article

alphaXiv Toggle

alphaXiv (What is alphaXiv?)

Links to Code Toggle

CatalyzeX Code Finder for Papers (What is CatalyzeX?)

DagsHub Toggle

DagsHub (What is DagsHub?)

GotitPub Toggle

Gotit.pub (What is GotitPub?)

Huggingface Toggle

Hugging Face (What is Huggingface?)

ScienceCast Toggle

ScienceCast (What is ScienceCast?)

Demos

Demos

Replicate Toggle

Replicate (What is Replicate?)

Spaces Toggle

Hugging Face Spaces (What is Spaces?)

Spaces Toggle

TXYZ.AI (What is TXYZ.AI?)

Related Papers

Recommenders and Search Tools

Link to Influence Flower

Influence Flower (What are Influence Flowers?)

Core recommender toggle

CORE Recommender (What is CORE?)

Author

Venue

Institution

Topic

About arXivLabs

arXivLabs: experimental projects with community collaborators

arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.

Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.

Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.

Which authors of this paper are endorsers? | Disable MathJax (What is MathJax?)

Key points and analysis

Article intelligence

InvestorsAdvanced

Key points

  • VeriOCRBench comprises 1,800 samples, with 1,600 trap-embedded invalid tasks covering 8 trap types and 4 verification dimensions, plus 200 trap-free controls.
  • The benchmark is built with a Visual Atomic Fact (VAF)-anchored pipeline and full human auditing, enabling decoupled evaluation of task verification, root-cause diagnosis, and over-refusal.
  • Evaluation of 15 leading MLLMs shows persistent blind compliance, diagnosis failures, and prompt-induced over-refusal.
  • The study emphasizes that models must verify whether the image premise, textual premise, and question jointly define an executable task before answering in OCR reasoning.

Highlights and analysis are generated automatically and may contain errors. Check the original source.