NTCIR

The Eighteenth NTCIR conference (NTCIR-18) would like to welcome all of you who are interested in and/or working on information access research. This sesquiannual event was first held in 1999, and has grown steadily ever since. About 303 people from 35 different countries attended NTCIR-16 and we are hoping that the event will be even more successful this time. NTCIR covers diverse information access tasks including, but not limited to, information retrieval, question answering, text summarisation and text mining, with an emphasis on East Asian languages and English.

At the NTCIR-18 conference, we will hear exciting reports from the following tasks:

- Automatic Evaluation of LLMs ("AEOLLM")
- The Second Fair Web Task ("FairWeb-2")
- Temporal Inference of Financial Arguments ("FinArg-2")
- Personal Lifelog Organisation & Retrieval Task ("Lifelog-6")
- Medical Natural Language Processing for AI Chat ("MedNLP-CHAT")
- Natural Language Processing for Radiology ("RadNLP")
- Resource Transfer Based Dense Retrieval ("Transfer-2")
- Hidden Causality Inclusion in Radiology Report Generation ("HIDDEN-RAD")
- Searching Unseen Sources for Historical Information ("SUSHI")
- Unifying, Understanding, and Utilizing Unstructured Data in Financial Reports ("U4")

While the NTCIR conference is a venue for task participants and organisers to discuss the official task results, open problems and future task designs, our door is open even to those who did not participate in any of the NTCIR-18 tasks. For the previous Conference, among them about half were researchers participated in the "task(s)" set by NTCIR, and others contributed to fruitful discussion.

NTCIR-18 will be co-located with EVIA 2025. EVIA is a one-day workshop for information access researchers to present their work on evaluation methods, measures, test collections, experimental designs and so on.

We look forward to seeing you all at NII in June!

* We will also provide online participation option. Details will be announced at a later date.