Mamba's per-word timesteps significantly predict human reading times beyond GPT-2 surprisal in a naturalistic dataset.
When High-Capacity Readers Slow Down and Low-Capacity Readers Speed Up: Working Memory and Locality Effects , volume =
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CL 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Timesteps of Mamba Align with Human Reading Times
Mamba's per-word timesteps significantly predict human reading times beyond GPT-2 surprisal in a naturalistic dataset.