RT @sebkrier: Ha! You can nerdsnipe reasoning models with decoy problems to make them overthink and slow them down/make them more expensive….

RT @ebagdasa: How Sudokus can waste your money? If you are using reasoning LLMs with public data, adversaries could pollute it with nonsens….

doing proof-based verification of outsourced machine learning workloads can be costly. A good workaround can be an ML model to detect maliciously modified predictions in real-time. Fides: A generative framework for outsourced ML workload validation via TEE.

RT @syhw: Do you need to quantize models? Try diffq, `pip install diffq` and

RT @WenhuChen: Tired of fine-tuning image generation models on each subject you care to generate? Today, we release SuTI, a zero-shot subje….