Rethinking Expressivity and Efficiency in Test-Time Training
Test-Time Training (TTT) enables long-context processing via continuous weight updates during inference, but current methods struggle to balance the expressivity of per-token update dynamics with the hardware efficiency of chunk-wise approximations. We propose E$^2$-TTT (Expressive and Efficient TTT) to bridge this gap. Under the standard approximation of taking gradients at the chunk-start weights, we derive a closed-form state transition that exactly reproduces the chunk-end fast-weight and momentum states of the per-token recurrence. This enables fully parallelized chunk-level training whil
Lineage graph
Paper → model → repo connections mined from source citations (Tier-1 exact match).
Why these links exist
Every edge carries a method, confidence, and the source snippet that justified it — so bad links are debuggable.
- PossiblePossibly related (embedding) · 52%Breakthrough in long-context efficiency announced →
- FuzzyOverlapping authors or contributors · 62%ultralytics/yolov5 →
“Shared author/contributor keys: martin”
- LinkedLinked via arxiv author · 85%Zeyun Zhong →
“Rethinking Expressivity and Efficiency in Test-Time Training”
- LinkedLinked via arxiv author · 85%Joya Chen →
“Rethinking Expressivity and Efficiency in Test-Time Training”
- LinkedLinked via arxiv author · 85%Manuel Martin →
“Rethinking Expressivity and Efficiency in Test-Time Training”
- LinkedLinked via arxiv author · 85%Frederik Diederichs →
“Rethinking Expressivity and Efficiency in Test-Time Training”
- LinkedLinked via arxiv author · 85%Juergen Gall →
“Rethinking Expressivity and Efficiency in Test-Time Training”
- LinkedLinked via arxiv author · 85%Juergen Beyerer →
“Rethinking Expressivity and Efficiency in Test-Time Training”