# A Test-time Actor-Critic Approach to News Images Generation

arXiv: https://arxiv.org/abs/2606.21304 · HTML: https://mlsanity.com/paper/2606.21304/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Placed in RL with diffusion and flow models, one of 15 directions in Deep reinforcement learning methods.
- Builds on 6 of its 12 references that are in the map.
- First posted on arXiv 19 Jun 2026.

## Placed in

- Deep reinforcement learning methods / [RL with diffusion and flow models](https://mlsanity.com/deeprl/rl-with-diffusion-and-flow-models/)

## Builds on (most cited first, 6 of 6)

- [Z-Image: An Efficient Image Generation Foundation Model with Single-Stream Diffusion Transformer](https://mlsanity.com/paper/2511.22699/) (2025)
- [Distribution Matching Distillation Meets Reinforcement Learning](https://mlsanity.com/paper/2511.13649/) (2025)
- [Prompting AI Art: An Investigation into the Creative Skill of Prompt Engineering](https://mlsanity.com/paper/2303.13534/) (2023)
- [PACE: Improving Prompt with Actor-Critic Editing for Large Language Model](https://mlsanity.com/paper/2308.10088/) (2023)
- [GenPilot: A Multi-Agent System for Test-Time Prompt Optimization in Image Generation](https://mlsanity.com/paper/2510.07217/) (2025)
- [Improving Calibration in Test-Time Prompt Tuning for Vision-Language Models via Data-Free Flatness-Aware Prompt Pretraining](https://mlsanity.com/paper/2604.27715/) (2026)

## Abstract (arXiv)

This paper introduces the CERTH-ITI solution for the MediaEval NewsImages 2026 challenge, which focuses on generating images related to news headlines. Inspired by the Actor-Critic paradigm in reinforcement learning, we present a test-time, model-agnostic Actor-Critic Image Generation approach (ACIG). ACIG generates prompts for image creation, produces the images, evaluates the generated results, and if needed refines the image generation prompts accordingly in a feedback loop. ACIG achieved the best results in the NewsImages 2026 challenge, according to the challenge's leaderboard.