@inproceedings{manoj-kumar-etal-2026-curiosai,
    title = "{C}urios{AI} at {S}em{E}val-2026 Task 8: Hybrid retrieval system with repeated sampling for generation",
    author = "Manoj Kumar, Aiswariya  and
      Takushima, Hiroki  and
      Beppu, Fumika  and
      Shibata, Yuki  and
      Yamaga, Daichi  and
      Hori, Takayuki",
    editor = "Kochmar, Ekaterina  and
      Ghosh, Debanjan  and
      North, Kai  and
      Komachi, Mamoru",
    booktitle = "Proceedings of the 20th {I}nternational {W}orkshop on {S}emantic {E}valuation (2026)",
    month = jul,
    year = "2026",
    address = "San Diego, California, USA",
    publisher = "Association for Computational Linguistics",
    url = "https://preview.aclanthology.org/ingest-acl-workshops/2026.semeval-1.151/",
    pages = "1106--1115",
    ISBN = "979-8-89176-414-9",
    abstract = "SemEval-2026 Task 8 (MTRAGEval) evaluates multi-turn Retrieval-Augmented Generation (RAG) under conversational challenges such as non-standalone turns, underspecification, and answerability detection. These conditions amplify retrieval and generation errors that standard single-turn RAG pipelines fail to address effectively. We present a robustness-oriented multi-turn RAG system combining contextual query rewriting, heterogeneous hybrid retrieval fused with Reciprocal Rank Fusion (RRF), domain-adaptive Low-Rank Adaptation (LoRA) reranking, and repeated sampling with metric-guided selection. On the official test set, our approach outperforms the organizers' baselines across all subtasks: Retrieval (nDCG@5: 0.5396 vs. 0.4795), Generation (0.7571 vs. 0.6390), and RAG (0.5486 vs. 0.5366). Our system ranks 5th in Subtask A, 5th in Subtask B, and 7th in Subtask C on the official leaderboard. These results demonstrate that calibrated hybrid retrieval combined with robust generation selection is effective for multi-turn RAG."
}