@inproceedings{liu-etal-2026-reflection,
    title = "Reflection in the Dark: Exposing and Escaping the Black Box in Reflective Prompt Optimization",
    author = "Liu, Shiyan  and
      Xia, Qifeng  and
      Xia, Qiyun  and
      Liu, Yisheng  and
      Yu, Xinyu  and
      Qu, Rui",
    editor = "T.Y.S.S., Santosh  and
      Rodriguez, Juan Diego  and
      de Gibert, Ona",
    booktitle = "Proceedings of the 64th Annual Meeting of the {A}ssociation for {C}omputational {L}inguistics ({ACL} 2026)",
    month = jul,
    year = "2026",
    address = "San Diego, California, United States",
    publisher = "Association for Computational Linguistics",
    url = "https://preview.aclanthology.org/ingest-acl/2026.acl-srw.8/",
    pages = "76--109",
    ISBN = "979-8-89176-393-7",
    abstract = "Automatic prompt optimization (APO) has emerged as a powerful paradigm for improving LLM performance without manual prompt engineering. Reflective APO methods such as GEPA iteratively refine prompts by diagnosing failure cases, but the optimization process remains black-box and label-free, leading to uninterpretable trajectories and systematic failure. We identify and empirically demonstrate four limitations: on GSM8K with a defective seed, GEPA degrades accuracy from 23.81{\%} to 13.50{\%}. We propose VISTA, a multi-agent APO framework that decouples hypothesis generation from prompt rewriting, enabling semantically labeled hypotheses, parallel minibatch verification, and interpretable optimization trace. A two-layer explore-exploit mechanism combining random restart and epsilon-greedy sampling further escapes local optima. VISTA recovers accuracy to 87.57{\%} on the same defective seed and consistently outperforms baselines across all conditions on GSM8K and AIME2025."
}