FDM achieves strictly O(1) decode memory via a fixed 272-slot cache while reaching 0.966 accuracy on multi-query associative recall, outperforming transformers by 59.5%.
MIPT-SSM : Scaling language models with O (1) inference cache via phase transitions
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Breaking the KV Cache Bottleneck: Fan Duality Model Achieves O(1) Decode Memory with Superior Associative Recall
FDM achieves strictly O(1) decode memory via a fixed 272-slot cache while reaching 0.966 accuracy on multi-query associative recall, outperforming transformers by 59.5%.