AI Paper Atlas

Research map / Vision-language and multimodal / Multimodal LLMs

TruthPrInt: Mitigating Large Vision-Language Models Object Hallucination Via Latent Truthful-Guided Pre-Intervention

Jinhao Duan, Fei Kong, Hao Cheng, James Diffenderfer, Bhavya Kailkhura, Lichao Sun, Xiaofeng Zhu, Xiaoshuang Shi, Kaidi Xu

ICCV 2025 · Vision-language and multimodal › Multimodal LLMs · approach: Decoding-time hallucination mitigation · 20 citations (OpenAlex)

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
apply TruthPrInt truthful-guided pre-interventionMiniGPT-4 greedy decoding · COCO CHAIRCHAIR_S: 29.53 -> 16.87Improved
apply TruthPrIntLlava-1.5 greedy decoding · COCO CHAIRCHAIR_S: 19.60 -> 10.33Improved
apply TruthPrIntmPLUG-Owl2 greedy decoding · COCO CHAIRCHAIR_S: 23.60 -> 11.13Improved
apply TruthPrIntMiniGPT-4 · COCO CHAIRBLEU: 15.58 -> 17.21Improved
apply TruthPrInt transferred from MiniGPT-4 detectorInternVL-2.5-8B · COCO CHAIRCHAIR_S: 13.2 -> 3.2Improved
apply TruthPrInt transferred from MiniGPT-4 detectorQwen2-VL-7B-Instruct · COCO CHAIRCHAIR_S: 12.0 -> 6.2Improved
apply TruthPrIntLlava-1.5 · Offline POPEPrecision: 92.80 -> 94.28Improved
apply TruthPrIntMiniGPT-4 · original POPE (avg)Precision: 60.74 -> 61.03No clear effect
increase max traceback N_B from 1 to 3 (ablation)TruthPrInt · COCO CHAIR (MiniGPT-4)CHAIR_S: 16.20 -> 15.20Improved
increase max traceback N_B from 3 to 5 (ablation)TruthPrInt · COCO CHAIR (MiniGPT-4)CHAIR_S: 15.20 -> 15.40No clear effect
use middle layer 14 instead of layer 12 hidden states (ablation)TruthPrInt · COCO CHAIR (MiniGPT-4)CHAIR_S: 24.20 -> 10.40Improved
use layer 20 instead of layer 14 hidden states (ablation)TruthPrInt · COCO CHAIR (MiniGPT-4)CHAIR_S: 10.40 -> 11.80Hurt

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 3 of 3 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