RESEARCH · RESEARCH · #2551
Test-Time Memory Evolution (TTME) outperforms Test-Time Training for scientific discovery
arXiv:2610.32046v1 introduces Test-Time Memory Evolution (TTME), a plug-and-play test-time adaptation method that updates an external memory (not model weights) by distilling experience during discovery. Evaluated under the same discovery framework as Test-Time Training (TTT), TTME outperforms TTT on five mathematical tasks while reducing FLOPs by 82.8–84.6% and runtime by 47.7–58.4%.
KEY POINTS
- arXiv:2610.32046v1 introduces Test-Time Memory Evolution (TTME), a plug-and-play test-time adaptation method that updates an external memory (not model weights) by distilling experience during discovery.
- Evaluated under the same discovery framework as Test-Time Training (TTT), TTME outperforms TTT on five mathematical tasks while reducing FLOPs by 82.8–84.6% and runtime by 47.7–58.4%.
- TTME provides a materially more compute-efficient alternative to TTT for discovery tasks, which could make adaptive methods practical at larger model scales.
WHY IT MATTERS
TTME provides a materially more compute-efficient alternative to TTT for discovery tasks, which could make adaptive methods practical at larger model scales.