Showing cs.LGShow all
2 papers · 1 filter
cs.LG2026
Latest Exact Match Attention
Moritz Brösamle
We introduce latest exact match attention (LEMA), an attention variant for transformers where queries and keys are binarized and each query attends only to the latest exactly match…
cs.LG2026
The Expressive Power of Low Precision Softmax Transformers with (Summarized) Chain-of-Thought
Moritz Brösamle, Stephan Eckstein
Existing expressivity results for transformers typically rely on hardmax attention, high precision, and other architectural modifications that disconnect them from the models used…