@HanGuo97: LLM training is built on fast MatMuls. But many surrounding ops still run as memory-bound kernels. CODA reparameterizes…
Summary
CODA reparameterizes memory-bound operations in LLM training to fuse them into the matmul epilogue, achieving near state-of-the-art performance with LLM-generated kernels.
View Cached Full Text
Cached at: 05/22/26, 05:49 AM
LLM training is built on fast MatMuls. But many surrounding ops still run as memory-bound kernels.
CODA reparameterizes them to hide in the matmul’s shadow, fused into its epilogue before results leave the chip.
Bonus: LLMs can write fast CODA kernels too (approaching SoLs). https://t.co/cOTeMUr4py
Similar Articles
Claude writing a macOS driver for my obscure HP printer built only for Windows
A user demonstrates using Claude AI to write a macOS driver for an HP printer that originally only supported Windows.
Do Uncertainty Signals Help? A Systematic Study of Uncertainty-Aware Decoding with Rollback Mechanisms
This paper systematically evaluates uncertainty-aware decoding with rollback mechanisms to improve code generation in large language models, demonstrating performance gains over standard methods using uncertainty signals.
Beyond Pass@k: Measuring Reliability and Security of Agentic Code Generation
This paper identifies critical flaws in current benchmarks for evaluating AI coding agents, such as the misapplication of the pass@k metric, and proposes new metrics like reliability@k and security-adjusted reliability@k to improve measurement of reliability and security.
When Uncertainty Isn't Enough: An Empirical Study of Self-Correction in Code Generation
The paper empirically studies self-correction in code generation using uncertainty estimation methods, finding that uncertainty-based approaches fail to improve Pass@1 accuracy, while verification-based methods yield significant gains.
Airbnb CEO Brian Chesky says AI writes 60% of its code—and sustaining ‘founder mode’ is the key to winning in the age of AI
Airbnb CEO Brian Chesky reveals that AI writes 60% of the company's new code and emphasizes 'founder mode' as crucial for businesses to adapt and thrive in the AI age, enhancing efficiency and innovation.