# Debiasing Text-to-Image Evaluation via Implicit Cultural Alignment Reward Modeling

arXiv: https://arxiv.org/abs/2607.15740 · HTML: https://mlsanity.com/paper/2607.15740/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Placed in Diffusion and multimodal model alignment, one of 10 directions in LLM alignment / RLHF / preference optimisation.
- Builds on 18 of its 33 references that are in the map.
- First posted on arXiv 17 Jul 2026.

## Placed in

- LLM alignment / RLHF / preference optimisation / [Diffusion and multimodal model alignment](https://mlsanity.com/alignment/diffusion-and-multimodal-model-alignment/)

## Builds on (most cited first, 18 of 18)

- [Attention Is All You Need](https://mlsanity.com/paper/1706.03762/) (2017)
- [High-Resolution Image Synthesis with Latent Diffusion Models](https://mlsanity.com/paper/2112.10752/) (2021)
- [Training language models to follow instructions with human feedback](https://mlsanity.com/paper/2203.02155/) (2022)
- [LoRA: Low-Rank Adaptation of Large Language Models](https://mlsanity.com/paper/2106.09685/) (2021)
- [Direct Preference Optimization: Your Language Model is Secretly a Reward Model](https://mlsanity.com/paper/2305.18290/) (2023)
- [Deep reinforcement learning from human preferences](https://mlsanity.com/paper/1706.03741/) (2017)
- [Improved Techniques for Training GANs](https://mlsanity.com/paper/1606.03498/) (2016)
- [CLIPScore: A Reference-free Evaluation Metric for Image Captioning](https://mlsanity.com/paper/2104.08718/) (2021)
- [Pick-a-Pic: An Open Dataset of User Preferences for Text-to-Image Generation](https://mlsanity.com/paper/2305.01569/) (2023)
- [Evaluating Text-to-Visual Generation with Image-to-Text Generation](https://mlsanity.com/paper/2404.01291/) (2024)
- [VIEScore: Towards Explainable Metrics for Conditional Image Synthesis Evaluation](https://mlsanity.com/paper/2312.14867/) (2023)
- [Towards Measuring and Modeling "Culture" in LLMs: A Survey](https://mlsanity.com/paper/2403.15412/) (2024)
- [ViSAGe: A Global-Scale Analysis of Visual Stereotypes in Text-to-Image Generation](https://mlsanity.com/paper/2401.06310/) (2024)
- [EVALALIGN: Supervised Fine-Tuning Multimodal LLMs with Human-Aligned Data for Evaluating Text-to-Image Models](https://mlsanity.com/paper/2406.16562/) (2024)
- [Evaluating Image Hallucination in Text-to-Image Generation with Question-Answering](https://mlsanity.com/paper/2409.12784/) (2024)
- [CulturalFrames: Assessing Cultural Expectation Alignment in Text-to-Image Models and Evaluation Metrics](https://mlsanity.com/paper/2506.08835/) (2025)
- [Culture in Action: Evaluating Text-to-Image Models through Social Activities](https://mlsanity.com/paper/2511.05681/) (2025)
- [CAIRe: Cultural Attribution of Images by Retrieval-Augmented Evaluation](https://mlsanity.com/paper/2506.09109/) (2025)

## Built on by (most cited first, 1 of 1)

- [When the Judge Acts: Auditing VLM-Guided Image Selection on Culturally Situated Prompts](https://mlsanity.com/paper/2610.01243/) (2026)

## Abstract (arXiv)

As Text-to-Image (T2I) systems rapidly advance, evaluating the cultural authenticity of synthesized content has become increasingly important for fair and trustworthy generative AI. Existing T2I evaluation metrics and multimodal judges often rely on visual-semantic representations that underrepresent implicit cultural norms, leading to biased preference judgments and the omission of fine-grained cultural cues. In addition, visual question answering (VQA)-based evaluators typically depend on autoregressive text generation, which limits their scalability for real-time reward modeling. To address these limitations, we introduce an Implicit Cultural Alignment Reward Model built upon a lightweight 4.2-billion-parameter Multimodal Large Language Model (MLLM). Our framework integrates an Implicit Cultural Probe with a Skip-connection Cross-Attention (SkipCA) mechanism, enabling late-stage semantic features to directly attend to early-stage visual representations and better preserve culturally salient details. Evaluations on 3,323 challenging and carefully curated image pairs from the CulturalFrames benchmark show that our approach achieves 83.49% pairwise accuracy, with Pearson and Kendall correlation coefficients of 0.5268 and 0.3749, respectively, outperforming representative vision-language metrics and MLLM-based evaluators. Moreover, by bypassing autoregressive text generation, our model processes each evaluation in 0.21 seconds under our local inference setup, achieving a 10× speedup over standard VQA-based evaluators. These results suggest that the proposed reward model can provide an efficient and culturally aware scalar signal for preference optimization pipelines such as Reinforcement Learning from Human Feedback and Direct Preference Optimization. Additional resources are available on our project page at https://bensonch1214.github.io/ImplicitCulturalAlignment/.