AI Paper Atlas

Research map / Image generation / Text-to-image diffusion

Culture in Action: Evaluating Text-to-Image Models through Social Activities

Sina Malakouti, Boqing Gong, Adriana Kovashka

ICLR 2026 · Image generation › Text-to-image diffusion · approach: Fairness and bias in generative models · 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
add hallucination (HAL) to alignment in the composite metric (ablation)FAITH with InternVL3 · GT-FAITH correlation (all)Spearman: 0.41 -> 0.44Improved
add exaggeration (EXAG) to ALIGN+HAL to form FAITH (ablation)FAITH with InternVL3 · GT-FAITH correlation (all)Spearman: 0.44 -> 0.47Improved
add exaggeration (EXAG) to ALIGN+HAL to form FAITH (ablation)FAITH with Qwen2.5-VL · GT-FAITH correlation (all)Spearman: 0.39 -> 0.42Improved
add HAL to ALIGN (ablation)FAITH with Qwen2.5-VL · GT-FAITH correlation (all)Spearman: 0.39 -> 0.39No clear effect
use descriptor-based FAITH instead of direct MLLM-as-judgeQwen2.5-VL · GT-FAITH correlation (all)Spearman: 0.10 -> 0.42Improved
use descriptor-based FAITH instead of direct MLLM-as-judgeInternVL3 · GT-FAITH correlation (all)Spearman: 0.20 -> 0.47Improved
use Proposer-Refiner reference descriptor generation instead of a single Proposer (GPT-4o) (ablation)reference descriptors · descriptor qualitySpearman: 0.28 -> 0.33Improved
use Proposer-Refiner instead of Gemini 2.5-Flash Proposer alone (ablation)reference descriptors · descriptor qualitySpearman: 0.30 -> 0.33Improved
raise similarity threshold tau calibrated on real images from 0.29 to 0.52 (ablation)ALIGN/HAL · correlation with humansSpearman: 0.21 -> 0.33Improved
evaluate prompts of Global South instead of Global North countriesQwen-Image · CULTIVateFAITH: 0.60 -> 0.55Hurt
evaluate Global South instead of Global North countriesQwen-Image · CULTIVateALIGN: 0.36 -> 0.30Hurt
evaluate Global South instead of Global North countriesQwen-Image · CULTIVateHAL: 0.51 -> 0.56Hurt

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 2 of 2 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