# PrismGPT: Proxy-Guided Learning for Region-Aware Photo Editing with Self-Synthesized Reasoning

arXiv: https://arxiv.org/abs/2609.24768 · HTML: https://mlsanity.com/paper/2609.24768/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Placed in Multimodal reasoning, one of 17 directions in LLM reasoning.
- Builds on 4 of its 30 references that are in the map.
- First posted on arXiv 21 Sep 2026.

## Placed in

- LLM reasoning / [Multimodal reasoning](https://mlsanity.com/lr/multimodal-reasoning/)

## Builds on (most cited first, 4 of 4)

- [Qwen3-VL Technical Report](https://mlsanity.com/paper/2511.21631/) (2025)
- [An Overview of Multi-Task Learning in Deep Neural Networks](https://mlsanity.com/paper/1706.05098/) (2017)
- [Uniworld-V2: Reinforce Image Editing with Diffusion Negative-aware Finetuning and MLLM Implicit Feedback](https://mlsanity.com/paper/2510.16888/) (2025)
- [PhotoArtAgent: Intelligent Photo Retouching with Language Model-Based Artist Agents](https://mlsanity.com/paper/2505.23130/) (2025)

## Abstract (arXiv)

Professional photo finishing relies on both global adjustments and region-specific local edits guided by semantic masks, yet current automated methods handle this workflow only partially. We present PrismGPT, a Vision-Language Model (VLM) framework that produces structured, region-aware editing plans from a single input image without relying on commercial black-box tools. Training a VLM to simultaneously diagnose aesthetic deficiencies at both global and local levels while predicting precise editing parameters is challenging due to the vast combinatorial decision space. We address this through proxy-guided learning: two simpler proxy tasks -- operation decomposition and region-aware aesthetic ranking -- teach the foundational skills the model needs, while a competence-based dynamic scheduler automatically rebalances the multi-task training ratio, progressively shifting emphasis from the proxy tasks to the primary editing task as each skill is mastered. Crucially, all reasoning traces used for supervised fine-tuning are self-synthesized by the same base model, eliminating the need for a stronger external teacher. Experiments on MIT-Adobe FiveK and SPIRE, a new professionally retouched benchmark we introduce, show that PrismGPT achieves state-of-the-art results while using only ~6% of the training data compared to the previous best method.