TY  - RPRT
TI  - Cross-Environment Neural Reranking for Sample-Efficient Action Selection in Text-Based Agents
AU  - Kan Shao
PY  - 2026
UR  - https://arxiv.org/abs/2606.02204
ID  - 2606.02204
ER  -