CATEGORICAL REPARAMETERIZATION WITH GUMBEL-SOFTMAX

https://arxiv.org/pdf/1611.01144

https://hanlab.mit.edu/blog/streamingllm

https://www.anthropic.com/engineering/demystifying-evals-for-ai-agents

"Attention", "Transformers", in Neural Network "Large Language Models"

http://bactra.org/notebooks/nn-attention-and-transformers.html

out of touch and embarrassingly misinformed

https://research.google/blog/alternating-updates-for-efficient-transformers/

https://matx.com/research/leaky_quantization?trk=feed_main-feed-card_reshare_feed-article-content

https://idlewords.com/2025/02/the_shape_of_a_mars_mission.htm#abort_code