On 24th July 2026, Ryotaro Shimada (B4) of the Department of Information and Communication Engineering and Lecturer Yusuke Matsui of the Department of Information and Communication Engineering, Graduate School of Information Science and Technology, received the Best Paper Award at the SynthIR Workshop, SIGIR 2026.
Best Paper Award, SynthIR Workshop at SIGIR 2026
This paper received the Best Paper Award, which is presented to the most outstanding research paper at "The First Workshop on Synthetic Content in Information Retrieval Ecosystems (SynthIR)," co-located with SIGIR 2026, a top international conference in the field of information retrieval.
About awarded research
Paper: Ryotaro Shimada, Yu-Chieh Lin, Yuji Nozawa, Youyang Ng, Osamu Torii, and Yusuke Matsui, "Towards Vision-Free CIR: Attribute-Augmented Scoring and LLM-Based Reranking for Zero-Shot Composed Image Retrieval"
In this study, we proposed a novel vision-free approach to Composed Image Retrieval that retrieves a target image from a reference image and textual instructions by representing images as text. By combining retrieval that considers image attributes with reranking using a large language model (LLM), our method outperforms existing zero-shot methods. This study was conducted as a collaborative research project with Kioxia Corporation.
https://zdzheng.xyz/SIGIR2026Workshop-SynthIR
Your impression & future
I would like to continue exploring new possibilities for multimodal retrieval, which is expected to be increasingly used in real-world applications.
Ryosuke Furuta, Department of Information and Communication Engineering, won a Suzuki Memorial Award in The Institute of Image Information and Television Engineers (ITE)
Rina Katsumata (D1), Department of Electrical Engineering and Information Systems, received WIT Student Research Encouragement Award
