An LLM agent can improve itself at test time by rewriting its surrounding executable harness from unlabeled traces, using only proxy signals and a frozen model.
International Conference on Learning Representations,
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.SE 1years
2026 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
TTHE: Test-Time Harness Evolution
An LLM agent can improve itself at test time by rewriting its surrounding executable harness from unlabeled traces, using only proxy signals and a frozen model.