ICML Workshop (2025)
Seohui Bae, Jeonghye Kim(KAIST), Youngchul Sung(KAIST), Woohyung Lim
Abstract
We introduce a test-time adaptive agent that performs exploratory inference through posterior-guided belief refinement without relying on gradient-based updates or additional training for LLM agent search operation under partial observability. Our agent maintains a structured belief over the environment state, iteratively updates it via action-conditioned observations and selects actions by maximizing predicted information gain over the belief space. We estimate information gain using a lightweight LLM-based surrogate and assess world alignment through a novel reward that quantifies the consistency between posterior belief and ground-truth environment configuration. Experiments show that our method outperforms inference-time scaling baselines such as prompt-augmented or retrieval-enhanced LLMs, in aligning with latent world states with significantly lower integration overhead.