Latent Lattice Language Models

E-Reverance 45 points 5 comments November 17, 2026
shivamduggal4.github.io · View on Hacker News

Discussion Highlights (5 comments)

augment_me

Seems like just an extension of Short-conv in the linear attention methods where we instead carry the EMA of the entire sequence instead of nearest 4 tokens. It would make sense in my opinion to do the evals based on the reason for needing Short-conv in the architectures that use it in the first place, where it was added to support exact retrieval/needle-in-haystack when we compress the attention state. It's odd to make this arch extension and not test beyond a 2k context window, or with the needle-in-haystack benchmark. In my head this should fail because the EMA cannot carry this granular information

royal__

How significant is a .98 decrease in perplexity. I'm a researcher in this field and I'm having trouble interpreting the implication of this.

tkslucas

Interesting approach to scale compute, never thought of using the per layer state like that

bionhoward

I am so excited for RNNs to be cool again

amitport

Isn't that "Infini-attention"? https://arxiv.org/abs/2404.07143 (2024)

Semantic search powered by Rivestack pgvector
13,334 stories · 126,150 chunks indexed