Future Corporation's Research Paper Accepted at ACL2026, a Leading International Conference in Natural Language Processing

A research paper co-authored by Yosuke Kishinami, Ryo Fujii, and Mutsumi Morishita—Research Engineers in the AI Strategy Promotion Group at Future Corporation (Shinagawa-ku, Tokyo; President and CEO: Tomohiko Taniguchi; hereinafter "Future")—has been accepted to the Main Conference of "ACL2026" (Association for Computational Linguistics 2026) (Note 1), a premier international conference in the field of natural language processing.

ACL2026 (The 64th Annual Meeting of the Association for Computational Linguistics) is a conference hosted by the Association for Computational Linguistics (ACL), the world's largest academic society in the field of natural language processing. The accepted paper will be presented at ACL2026, to be held in San Diego, USA, from July 2 to July 7, 2026. This research is a collaborative effort with Nara Institute of Science and Technology, a national university corporation.

Accepted Paper

Revisiting Non-Verbatim Memorization in Large Language Models: The Role of Entity Surface Forms Yuto Nishida, Naoki Shikoda, Yosuke Kishinami*, Ryo Fujii*, Makoto Morishita*, Hidetaka Kamigaito, Taro Watanabe *Affiliated with Future Corporation

https://aclanthology.org/2026.acl-long.2178.pdf

【Research Overview】

Large language models (LLMs) store various kinds of knowledge. This study investigates whether knowledge memorized by LLMs can be equally retrieved regardless of the name used—such as official names, aliases, or abbreviations. To evaluate this, we constructed a new evaluation dataset called RedirectQA, using alternative mentions of people and organizations from Wikipedia, and tested multiple LLMs. The results show that LLM responses often vary depending on the name used to refer to the same entity. This reveals that robust evaluation of LLMs must consider their ability to handle diverse naming variations.

At Future, we are actively cultivating and hiring research engineers and AI engineers, primarily through our specialized AI organization, the AI Strategy Promotion Group, advancing academic research and development in natural language processing and generative AI. In 2024, with support from GENIAC (Note 2), a national project to strengthen domestic generative AI development, we publicly released a foundational model specialized in the Japanese language and software development. Starting in 2025, we launched the "Future PhD Support Program" (Note 3), a corporate doctoral support initiative to assist employees in obtaining PhDs, accelerating the development of talent in AI and other advanced technologies. Furthermore, together with our core operating company Future Architect (Headquarters: Shinagawa-ku, Tokyo; President and CEO: Tomohiko Taniguchi), we are accelerating AI adoption to support clients' management and business operations, aiming to become Japan's number one company in AI social implementation (Note 4).

Moving forward, we will continue to holistically design clients' operations and systems through cutting-edge research and scientific consulting approaches, creating new value.

Note 1. Official website of "ACL2026": https://2026.aclweb.org/

Note 2. GENIAC https://www.meti.go.jp/policy/mono_info_service/geniac/index.html

NEDO-funded project https://www.nedo.go.jp/koubo/IT2_100331.html

URL for the publicly released "Foundational Model Specialized in Japanese and Software Development"

https://huggingface.co/future-architect/Llama-3.1-Future-Code-Ja-8B

Note 3. Future Introduces Corporate Doctoral Support Program "Future PhD Support Program"

https://www.future.co.jp/press_room/PDF/PressRelease_FuturePhDSupportProgram_250423.pdf

Note 4. Future Announces Vision to Become Japan's No.1 Company in AI Social Implementation

Accelerating Social Problem Solving Through "AI-Driven Development" with 3x Productivity and AI×IT Total Design https://www.future.co.jp/press_room/PDF/PressRelease_FUTURE_AIstatement_20260430.pdf