AI Paper Atlas

Research map / Vision-language and multimodal / Robot action and navigation

Forensic Prompting with Dual-Action Policy Optimization for Vision-Language Forgery Detection and Localization

Ye Zhu, Ai Zhao, Jinwei Wang

ICML 2026 · Vision-language and multimodal › Robot action and navigation · approach: Vision-language-action foundation models

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
add Forensic Prompting Module (prompt bank + router) (ablation)FPDA base · CocoGlide detectionAUC: 0.905 -> 0.925Improved
add DAPO image-side action (ablation)FPDA base · CocoGlide localizationF1: 0.690 -> 0.730Improved
combine FPM and DAPO (ablation)FPDA · CocoGlide detectionAUC: 0.925 -> 0.941Improved
add Prompt Bank (ablation)FPM · SID Set description detectionAUC: 0.985 -> 0.989Improved
add Prompt Router (ablation)FPM · SID Set description detectionAUC: 0.989 -> 0.993Improved
fuse optional text without reliability gating (ablation)FPM bank+router · SID Set description detectionAUC: 0.993 -> 0.992Hurt
add reliability gate to text fusion (ablation)FPM · SID Set description detectionAUC: 0.992 -> 0.996Improved
enable text-side action (ablation)DAPO · NIST16 detectionAUC: 0.831 -> 0.863Improved
enable image-side action (ablation)DAPO · NIST16 detectionAUC: 0.831 -> 0.871Improved
use both text and image actions (ablation)DAPO · NIST16 detectionAUC: 0.871 -> 0.902Improved
add detection reward (ablation)DAPO · CASIA v1 detectionAUC: 0.953 -> 0.975Improved
add localization reward (ablation)DAPO · CASIA v1 localizationF1: 0.702 -> 0.738Improved

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy