ChatSearch: a Dataset and a Generative Retrieval Model for General Conversational Image Retrieval
Keywords: Image Retrieval; Multimodal Learning; Conversational Image Retrieval; Human-computer Interaction
Abstract: In this paper, we investigate the task of general conversational image retrieval on open-domain images.
The objective is to search for images based on interactive conversations between humans and computers. To advance this task, we curate a dataset called ChatSearch. This dataset includes a multimodal conversational context query for each target image, thereby requiring the retrieval system to infer the underlying retrieval intention from the multimodal dialogue conducted over multiple rounds.
Simultaneously, we propose a generative retrieval model named ChatSearcher, which is trained end-to-end to accept and produce interleaved image-text inputs/outputs. ChatSearcher exhibits strong capability in reasoning with multimodal context and can leverage world knowledge to yield more sophisticated retrieval results. It demonstrates superior performance on the ChatSearch dataset and also achieves competitive results on other image retrieval tasks, such as zero-shot text-to-image retrieval and zero-shot composed image retrieval. With the availability of the ChatSearch dataset and the effectiveness of the ChatSearcher model, we anticipate that this work will inspire further research on interactive multimodal retrieval systems.
Primary Area: datasets and benchmarks
Code Of Ethics: I acknowledge that I and all co-authors of this work have read and commit to adhering to the ICLR Code of Ethics.
Submission Guidelines: I certify that this submission complies with the submission instructions as described on https://iclr.cc/Conferences/2024/AuthorGuide.
Anonymous Url: I certify that there is no URL (e.g., github page) that could be used to find authors' identity.
No Acknowledgement Section: I certify that there is no acknowledgement section in this submission for double blind review.
Submission Number: 659
Loading