AI Paper Atlas

Research map / Vision-language and multimodal / Multimodal LLMs

TrustGen: A Platform of Dynamic Benchmarking on the Trustworthiness of Generative Foundation Models

Yue Huang, Chujie Gao, Siyuan Wu, Haoran Wang, Xiangqi Wang, Jiayi Ye, Yujun Zhou, Yanbo Wang, Jiawen Shi, Qihui Zhang, Han Bao, Zhaoyi Liu and 52 more

ICLR 2026 · Vision-language and multimodal › Multimodal LLMs · approach: Interleaved image-text generation with MLLMs · 0 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted summary. Written by AI from the paper's full text. Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Introduces TrustGen, a platform for dynamic benchmarking of the trustworthiness of generative foundation models, addressing fragmented and outdated existing evaluations.

Problem

Assessing the trustworthiness of generative foundation models (LLMs and text-to-image systems) is critical but existing evaluation efforts are fragmented, quickly outdated, and lack extensibility.

Approach

A dynamic benchmarking platform for trustworthiness of GenFMs intended to stay current and extensible, rather than a static fixed benchmark suite.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
use open-source CogView-3-Plus instead of proprietary DALL-E-3text-to-image model · TrustGen T2I averagetrustworthiness score: 72.52 -> 74.96Improved
use proprietary DALL-E-3 instead of open-source T2I modelstext-to-image model · T2I Truthfulnesstrustworthiness score: 44.80 vs 28.06-35.67 for open modelsImproved
use proprietary vs open-source LLM (Llama-3.2-70B vs GPT-4o)LLM · TrustGen LLM trustworthinesstrustworthiness scoreNo clear effect
scale up VLM size (Llama-3.2-11B-V to 90B-V)VLM · TrustGen VLM averagetrustworthiness score: 64.91 -> 47.33Hurt
larger LLM size within a seriesLLM · LLM privacyprivacy preservationHurt
larger LLM size within a seriesLLM · LLM fairnessfairness metricsImproved
larger VLM sizeVLM · VLM privacyprivacy scoreMixed
apply input perturbationsLLMs and VLMs · robustnessperformanceMixed
evaluate on dynamically generated vs established benchmark dataLLMs · truthfulnessperformanceImproved
use proprietary instead of open-source VLMsVLM · jailbreak safetyrefusal-to-answer rateImproved

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