InternLM 开源 ETCHR：面向 MLLM 的解耦式视觉推理图像编辑器
GitHub - InternLM/ETCHR: A question-conditioned, reasoning-aware image editor designed to serve as a decoupled visual reasoning assistant for Multimodal Large Language Models (MLLMs). · GitHub
InternLM 发布 ETCHR（Editing To Clarify and Harness Reasoning），一个以问题为条件、具备推理感知能力的图像编辑器，作为多模态大语言模型（MLLM）的解耦式视觉推理助手，可即插即用地辅助 Qwen3-VL-8B、Gemini-3.1-Flash-Lite、Kimi K2.5 等下游模型而无需微调。其 Edit-Verify-Reason 推理机制在验证失败时回退原图，在细粒度感知、图表理解、逻辑推理、拼图还原和 3D 理解五类任务上均提升 Pass@1 准确率。项目已开源 ETCHR-FLUX.2-klein-9B 模型、SFT-400K 与 GRPO-10K 数据集及 DL3DV-2K 基准，基于 FLUX.2-klein-base-9B，遵循 FLUX 非商用许可。
正文
当前显示原文，中文全文译文尚未就绪。
InternLM; ETCHR
Repository navigation
NotificationsYou must be signed in to change notification settings; Fork0 (0); Star23 (23)You must be signed in to star a repository
A question-conditioned, reasoning-aware image editor designed to serve as a decoupled visual reasoning assistant for Multimodal Large Language Models (MLLMs).
Repository files navigation
README; License
*Equal Contribution †Corresponding authors.
📖Paper | 🏠Homepage | 🤗ETCHR-FLUX.2-klein-9B Model | 🤗ETCHR SFT-400K Dataset | 🤗ETCHR GRPO-10K Dataset | 🤗DL3DV-2K Benchmark
🚀 [2026/05/24] We have released the training and evaluation code of ETCHR; 🚀 [2026/05/21] We have released the ETCHR-FLUX.2-klein-9B Model, ETCHR-SFT-400K Dataset and ETCHR GRPO-10K Dataset
🌈 Overview
We are thrilled to introduce ETCHR (Editing To Clarify and Harness Reasoning), a novel question-conditioned, reasoning-aware image editor designed to serve as a decoupled visual reasoning assistant for Multimodal Large Language Models (MLLMs).
By decoupling the specialized image editor from the downstream understanding model, ETCHR bridges the critical bottleneck where a purely textual chain of thought fails in fine-grained focus or complex spatial transformations.
💡 Highlights
🔥 Decoupled & Plug-and-Play: ETCHR functions as a separate module, allowing it to assist diverse downstream MLLMs (such as Qwen3-VL-8B, Gemini-3.1-Flash-Lite, or Kimi K2.5) without requiring any task-specific fine-tuning on the understanding models themselves; 🔥 Naturally Reflective Pipeline: Introduces an Edit-Verify-Reason inference mechanism where the understanding model filters out noisy or flawed edits, reverting safely to the original image when verification fails
📊 Results
We evaluate ETCHR across five distinct task families spanning fine-grained perception, chart understanding, logic reasoning, jigsaw restoration, and 3D understanding. Across all evaluated backbones, ETCHR consistently yields major improvements in Pass@1 accuracy:
🛠️ Evaluation
Prepare your environment:
We Provide an example code running ETCHR on DL3DV-2K Benchmark in Evaluation/inference_dl3dv.py, you can start the evaluation with the following two steps:
Step 1: start a VLLM server for an understanding model (eg. Qwen3-VL-8B, Kimi K2.5, ...).
Step 2: Run ETCHR atop any understanding model
🛠️ Training
We adopt a two-stage Training Pipeline. See SFT.md and RL.md for further details.
Cases
ETCHR can assist with a broad spectrum of understanding tasks, including fine-grained perception, chart reasoning, maze navigation, jigsaw puzzles, and 3D spatial understanding.
✒️Citation
If you find this project useful, please kindly cite:
📄 License
Our work is based on FLUX.2-klein-base-9B, so please follow FLUX Non-Commercial License.
❤️ Acknowledgement
The work is built upon DiffSynth-Studio and Pref-GRPO, two excellent codebases for Diffusion models training!
A question-conditioned, reasoning-aware image editor designed to serve as a decoupled visual reasoning assistant for Multimodal Large Language Models (MLLMs).