EviScene: Beyond "Is It AI?" for Agentic Image Factuality Assessment

Hugging Face Daily Papers Papers
multimodal-agent factuality benchmark image-verification retrieval-augmented llm-agent evidence-grounding

Summary

该论文提出Image Factuality Assessment任务及EviScene多模态ReAct智能体，通过视觉检查、外部检索与证据分析的交替推理来判定图像所描绘场景是否事实成立，并发布了包含10,174张图像的EviLens专家审阅基准，在Qwen3.5-9B上实现了79.05%的平衡准确率。

Determining whether an image is AI-generated does not establish whether the situation it depicts is factually supported. We formulate Image Factuality Assessment: given an image and a generic verification instruction, a system returns a supported or refuted verdict with an evidence-grounded explanation. Without a supplied textual claim, the system must progressively refine factual questions during investigation and align external evidence with the specific situation depicted. We introduce EviLens, an expert-reviewed benchmark of 10,174 supported and refuted images from controlled construction and web collection, covering twelve factual error types and evaluating both verdict accuracy and evidence sufficiency. We develop EviScene, a multimodal ReAct agent that interleaves visual inspection, external retrieval, and evidence analysis, using new findings to refine its questions and revisit relevant image details. We train a compact policy with supervised fine-tuning on reviewed teacher trajectories, followed by privileged on-policy self-distillation applied to verified decisions from the policy's own investigations. On the 1,527-image test set, the trained Qwen3.5-9B agent achieves 79.05% balanced accuracy and a Strict Evidence Sufficiency Rate (SESR) of 52.26%, improving over the same-backbone base agent by 6.71 and 9.37 percentage points, respectively. Its SESR is the highest among the evaluated agents.Code, models, and data will be released at https://github.com/wangzian828/EVISCENE and https://huggingface.co/ZiAnwang/EviScene.
Original Article
View Cached Full Text

Cached at: 10/14/26, 10:53 AM

Paper page - EviScene: Beyond “Is It AI?” for Agentic Image Factuality Assessment

Source: https://huggingface.co/papers/2610.14609 Published on Oct 10

·

Submitted byhttps://huggingface.co/Zhengbo-Zhang

Boon Oct 14

Authors:

,

,

,

,

,

,

,

,

,

,

,

Abstract

DeterminingwhetheranimageisAI-generateddoesnotestablishwhetherthesituationitdepictsisfactuallysupported.WeformulateImageFactualityAssessment:givenanimageandagenericverificationinstruction,asystemreturnsasupportedorrefutedverdictwithanevidence-groundedexplanation.Withoutasuppliedtextualclaim,thesystemmustprogressivelyrefinefactualquestionsduringinvestigationandalignexternalevidencewiththespecificsituationdepicted.WeintroduceEviLens,anexpert-reviewedbenchmarkof10,174supportedandrefutedimagesfromcontrolledconstructionandwebcollection,coveringtwelvefactualerrortypesandevaluatingbothverdictaccuracyandevidencesufficiency.WedevelopEviScene,amultimodalReActagentthatinterleavesvisualinspection,externalretrieval,andevidenceanalysis,usingnewfindingstorefineitsquestionsandrevisitrelevantimagedetails.Wetrainacompactpolicywithsupervisedfine-tuningonreviewedteachertrajectories,followedbyprivilegedon-policyself-distillationappliedtoverifieddecisionsfromthepolicy’sowninvestigations.Onthe1,527-imagetestset,thetrainedQwen3.5-9Bagentachieves79.05%balancedaccuracyandaStrictEvidenceSufficiencyRate(SESR)of52.26%,improvingoverthesame-backbonebaseagentby6.71and9.37percentagepoints,respectively.ItsSESRisthehighestamongtheevaluatedagents.Code,models,anddatawillbereleasedathttps://github.com/wangzian828/EVISCENEandhttps://huggingface.co/ZiAnwang/EviScene.

View arXiv pageView PDFAdd to collection

Get this paper in your agent:

hf papers read 2610\.14609

Don’t have the latest CLI?curl \-LsSf https://hf\.co/cli/install\.sh \| bash

Models citing this paper1

#### ZiAnwang/EviScene Updatedabout 2 hours ago • 6

Datasets citing this paper0

No dataset linking this paper

Cite arxiv.org/abs/2610.14609 in a dataset README.md to link it from this page.

Spaces citing this paper0

No Space linking this paper

Cite arxiv.org/abs/2610.14609 in a Space README.md to link it from this page.

Collections including this paper0

No Collection including this paper

Add this paper to acollectionto link it from this page.

Similar Articles

EviRover: Reinforcing Agentic Perception Beyond a Glance

Hugging Face Daily Papers

EviRover reframes visual perception as an agentic process that gathers evidence beyond a single glance, training a 4B perception agent with SFT and agentic RL on newly generated datasets (EviRover-SFT-5K, EviRover-RL-12K) and a human-verified benchmark (EviLens), outperforming its backbone by 30 points and rivaling proprietary models.

Reasoning-Informed Visual Editing

Hugging Face Daily Papers

该论文提出了 RISEBench 及其扩展版本 RISEBench++，这是首个评估推理驱动视觉编辑（Reasoning-Informed Visual Editing）的基准，涵盖时序、因果、空间、逻辑、反事实及混合推理等六个维度，并提出训练无关的 RISE-Agent 框架，结果显示即使是 GPT-Image-2.5 Sunburst 也仅达 56.6% 准确率。

OneSearch-VL: Unified Multimodal Deep Research Agent for Image and Video

Hugging Face Daily Papers

该论文提出 OneSearch-VL，一个以视觉接地证据图（VGEG）为核心的统一多模态深度研究智能体，用于单图、多图和视频研究任务，并构建了 SFT/RL 数据集及细粒度评测基准，在多个图文与视频基准上显著超越 Qwen3-VL-8B。