clear3 min read
CLEAR: Legible and Steerable Recommendation via Grounded Natural-Language Interests
Modern recommenders keep learned preferences in representations people cannot inspect or change. We present CLEAR (Controllable and Legible End-to-end Architecture for Recommendation), an end-to-end system deployed at Meta and built around an editable natural-language interest layer.
By Qi Guo, Reza Mirghaderi, Iordanis Fostiropoulos, David Fan, Yuchen Liu, Shiliang Tang, Lukasz Korycki, Golnaz Ghasemiesfeh, Xinzhu Bei, Sasha Vorontsov, Hanchao Yu, Brendan Quinlivan, Khoi Pham, Jigar Gada, Hao Yu, Felix Li, Hang Gao, Boyu Fang, Yang Gao, Yale Thomas, Xiaomeng Jin, Avishek Dutta, Hong Hong, Janet Xia, Ethan Zhu, Ellie Pan, Xiang Shen, Jawwad Zakaria, Jinfeng Lin, Jonathan Rush, Ruixue Lian, Noah Kennedy, Brian Vanover, Jas Pyneni, Daniel Thiyagu, Apoorv Saxena, Sanyogeeta LawandarXiv - cs.AI

Read the full story
Originally published by arXiv - cs.AI on Nov 17, 2026.
Comments
Sign in to join the discussion.