arxiv
PublishedJuly 2, 2026 at 4:00 AM
Beyond Perplexity: A Behavioral Evaluation Framework for Deployment-Memory Claims in LLM Test-Time Training
Publisher summary· verbatim
arXiv:2607.00368v1 Announce Type: new Abstract: Large language model test-time training (TTT) is often evaluated through local proxy metrics: models are updated on recent tokens, retrieved context, target-domain data, or verifiable task attempts, and then judged by perplexity, future-token loss, lon
Stay posted· Newsletter
A 5-min weekly brief — top movers, price watch, story of the week.
Discussion
No replies yet. Be first.
Related coverage
More from ARXIV
arxivFAD-SA-GRU: Enhancing Hate Speech Detection in Algerian Dialect Through Feature-Augmented Self-Attention GRU Networks11harxivSMETA-ZSL:Semantic Meta-Alignment for Zero-Shot Threat Classification11harxivDzongkha Next Word Prediction System11harxivMoP-JEPA: Hard-Assigned Predictor Mixtures for Stochastic JEPA World Models11hThe Bubble Brief
WEEKLYRead AI insights every Tuesday — top movers, new releases, story of the week.
Originally published on arxiv ↗