🛰️ Daily AI Frontier
‹ back to 2026-08-23

Rethinking Expressivity and Efficiency in Test-Time Training

Research Efficiency & Systems

Ranking

Overall 90
Content 100
Popularity 66

Observed public metrics from 1 member.

Representative image for Rethinking Expressivity and Efficiency in Test-Time Training

Merged summary

TL;DR - E²-TTT is a test-time training method that parallelizes chunk-level computation while preserving the fast-weight and momentum states of per-token updates. It combines the throughput of chunk-wise methods with stronger long-context retrieval and length extrapolation.

  • Derives a closed-form chunk transition that exactly reproduces per-token recurrence states under the chunk-start gradient approximation.
  • Trains models up to 1.3B parameters from scratch and matches prior TTT and hybrid-attention baselines in language modeling.
  • Outperforms those baselines on in-context retrieval.
  • Retains over 90% passkey accuracy at eight times its training context length while matching efficient chunk-wise training throughput.

Sources (1)

Rethinking Expressivity and Efficiency in Test-Time Training

arXiv cs.LG Zeyun Zhong, Joya Chen, Manuel Martin, Frederik Diederichs, Juergen Gall, Juergen Beyerer 2026-08-21 arXiv:2608.21308
Public signals Hugging Face upvotes 1
Providers: Hugging Face · Upvotes 1 OpenAlex · N/A Publisher · N/A Semantic Scholar · N/A X · N/A Fetched 2026-09-22 14:33:08.287018 UTC

TL;DR - E²-TTT is a test-time training method that parallelizes chunk-level computation while preserving the fast-weight and momentum states of per-token updates. It combines the throughput of chunk-wise methods with stronger long-context retrieval and length extrapolation.

  • Derives a closed-form chunk transition that exactly reproduces per-token recurrence states under the chunk-start gradient approximation.
  • Trains models up to 1.3B parameters from scratch and matches prior TTT and hybrid-attention baselines in language modeling.
  • Outperforms those baselines on in-context retrieval.
  • Retains over 90% passkey accuracy at eight times its training context length while matching efficient chunk-wise training throughput.
item →