Tag
Scientists are adopting 'vibe coding', a conversational approach using AI tools like LLMs to generate code for data visualization and analysis, speeding up research but requiring caution.
This paper introduces GRAFT-ATHENA, a self-improving agentic framework that autonomously discovers and evolves numerical algorithms for scientific problems. It demonstrates near-machine-precision accuracy on physics-informed machine learning benchmarks and successfully tackles complex engineering challenges.
This paper introduces Christoffel-DPS, a distribution-free framework for optimal sensor placement in diffusion posterior sampling that outperforms classical Gaussian-based methods. It provides theoretical guarantees and practical improvements for reconstructing states from complex, non-Gaussian distributions using generative models.
Metal-Sci introduces a 10-task benchmark for optimizing scientific computing kernels on Apple Silicon, paired with an evolutionary search framework driven by large language models. The study evaluates models like Claude Opus 4.7, Gemini 3.1 Pro, and GPT 5.5, demonstrating significant speedups while using out-of-distribution testing to catch silent performance regressions.