@inproceedings{sun-etal-2026-beyond-query,
    title = "Beyond Query Bias: Candidate-Aware Iterative Refinement for Zero-Shot Composed Image Retrieval",
    author = "Sun, Nan  and
      Tang, Jing  and
      Sun, Lei  and
      Chen, Rui  and
      Lu, Yuxing  and
      Chu, Xiangxiang  and
      Ling, Hefei  and
      Cai, Yujun",
    editor = "Liakata, Maria  and
      Moreira, Viviane P.  and
      Zhang, Jiajun  and
      Jurgens, David",
    booktitle = "Findings 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.findings-acl.1120/",
    pages = "22318--22329",
    ISBN = "979-8-89176-395-1",
    abstract = "Zero-Shot Composed Image Retrieval (ZS-CIR) retrieves target images using a reference image and modification text without task-specific training. Existing methods typically rely on MLLMs to generate query vectors with pre-trained models like CLIP. However, those constructed queries suffer from inherent cognitive bias due to unknown candidate distribution. We propose CoRR, a training-free framework that reframes ZS-CIR as a self-correcting process through bias-aware query refinement. CoRR uses retrieved results as feedback to perceive the candidate distribution. With carefully designed CoT prompting, the MLLM inspects the retrieved candidates to identify intent misalignments in the query and then corrects them via Historical Query Fusion. We also introduce Retrieval-Driven Caption Optimization to provide context-aligned examples, reducing phrasing and style mismatches. Experiments on public benchmarks show that CoRR significantly outperforms other SOTA methods."
}