@MaximeRivest: Compound AI System for Images are way under appreciated. We need gepa, dspy, autoresearch style optimization to go from…
Summary
Maxime Rivest argues that compound AI systems for images are undervalued and suggests leveraging optimization frameworks like DSPy and GEPA to automate pipeline creation involving SAM and classifiers.
Similar Articles
@MaximeRivest: https://x.com/MaximeRivest/status/2055293570119065875
MaximeRivest explains DSPy's five core components—Optimizers, Signatures, LMs, Modules, and Adapters—and argues that effective AI engineering requires mastering these elements, highlighting the often-overlooked role of rendering structured outputs.
@MaximeRivest: This is my type of compound AI systems! Beautiful.
Wuli-Art introduces Gemma-4-for-Qwen-Image-Edit-2511-Prompt-Extend, a prompt optimizer that enhances image editing prompts for improved precision and clarity.
@MaximeRivest: There is still so much to do! We know deep learning works, we just need to be a little more sharing and organized! I sp…
The author discusses the need for better sharing and organization in deep learning to make it more accessible to biologists for automating tasks like image analysis and measurement.
@harold_matmul: dspy.GEPA used in pretraining data curation in the new Microsoft AI effort :-)
The article explains how GEPA (Genetic-Pareto Optimization) within DSPy is used for efficient prompt tuning, specifically applied to pretraining data curation at Microsoft AI, allowing researchers to replace manual prompt engineering with automated compute-driven optimization.
@aiDotEngineer: Building Generative Image & Video models at Scale https://youtube.com/watch?v=xOP1PM8fwnk… A lot of interest in image g…
YouTube talk by @sedielem offering a concise state-of-the-art overview of scaling generative image and video models, covering modeling, architecture, distillation and control.