AI Paper Atlas

Research map / Vision-language and multimodal / Multimodal LLMs

Delineating Knowledge Boundaries for Honest Large Vision-Language Models

Junru Song, Yimeng Hu, Yijing Chen, Huining Li, Qian Li, Lizhen Cui, Yuntao Du

ECCV 2026 · Vision-language and multimodal › Multimodal LLMs · approach: VLM capability analysis · 2 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
align with CPO on Visual-Idk preference pairs (few-shot)LLaVA-1.5 7B · V-Idk testTruthful rate: 57.9 -> 67.3Improved
SFT-only refusal tuning (ablation)LLaVA-1.5 7B · V-Idk test (zero-shot)IK-IK rate: 45.8 -> 34.4Hurt
SFT-only refusal tuning (ablation)LLaVA-1.5 7B · V-Idk test (zero-shot)Truthful rate: 47.6 -> 67.1Improved
use preference optimization (ORPO) instead of SFT (ablation)LLaVA-1.5 13B · V-Idk test (few-shot)Truthful rate: 67.3 (SFT) -> 76.6 (ORPO)Improved
use DPO instead of SFT/CPO/ORPO (ablation)LLaVA-1.5 7B · V-Idk test (zero-shot)Truthful rate: 47.6 -> 57.4No clear effect
add few-shot demonstrations at inference (ablation)LLaVA-1.5 7B base · V-Idk testTruthful rate: 47.6 -> 57.9Improved
scale model from 7B to 13B (ablation)LLaVA-1.5 ORPO · V-Idk test (zero-shot)Truthful rate: 60.0 -> 66.2Improved
apply V-Idk-trained ORPO to ScienceQALLaVA-1.5 7B · ScienceQA (few-shot)Truthful rate: 56.1 -> 63.8Improved
apply DPO to OOD scenarios (ablation)LLaVA-1.5 7B · VizWiz-Unans / PMC-VQATruthful (refusal) rate: falls to 82.8% on VizWizHurt
SFT on blurry VizWiz-Unans (ablation)LLaVA-1.5 7B · VizWiz-UnansTruthful (refusal) rate: 95.6 (+11.3 over baseline)Improved
CPO on medical PMC-VQA (ablation)LLaVA-1.5 7B · PMC-VQATruthful (refusal) rate: 93.9Improved
ORPO alignment changes refusal Logprob for unknown questionsLLaVA-1.5 7B · Visual-Idk probingLogprob: -0.80 -> -0.63Improved

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