Reproduce “CLEAR: Legible and Steerable Recommendation via Grounded Natural-Language Interests” on OpenResearch

Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.

Universal app for Apple silicon and Intel. Requires macOS 11 or later.

About this paper

Qi Guo, Reza Mirghaderi, Iordanis Fostiropoulos, David Fan, Yuchen Liu, Shiliang Tang, Lukasz Korycki, Golnaz Ghasemiesfeh and 46 more

CLEAR links an editable interest identity to recommendation serving, so users can inspect interests and add or suppress them. It grounds inferred interests in users’ histories, while additions can retrieve topical candidates without prior support. In review, users rated 88.5% of 7,261 displayed interests accurate; a randomized cumulative ablation found sourcing’s removal reduced same-day matching delivery by 13.7 percentage points in that sequence.

Summary by alphaXiv.

Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.