cs.CVNov 19, 2026

Beyond Covers and Titles: Evidence-First Ranking for Robust Cold-Start Recommendation

Authors: Fuyi Wang, Yuxin Cao, Yifan Liao, Renyang Liu, Lin Li, Wei Song

Organizations: Royal Melbourne Institute of Technology, Australia · National University of Singapore, Singapore · The Hong Kong University of Science and Technology (Guangzhou), China · Southern Cross University, Australia · 

Abstract

Short-video platforms rank new uploads before anyone has watched them, using content towers that read a video's cover, title and frames. Existing attacks on content-aware recommenders perturb pixels or features and ignore platform re-encoding, and existing defenses train against norm-bounded perturbations that neither match real edits nor preserve recommendation quality. We show with REAP, a realizable attack through {a simulated platform channel (H.264 transcoding, rescaling, cropping and JPEG)}, that packaging edits raise a cold video by 32.0 rank-percentile points in unrelated audiences. We then observe that cold-start ranking does not need the packaging at all: a tower that sees only the video body ranks cold items best. Building on this, we propose Evidence-First Ranking (EFR), which scores cold videos from their body, admits packaging only through a support-gated residual bounded by γγ, and merges cold and warm rankings by rank percentile, yielding an exact certificate against any packaging edit. On MicroLens-100K, EFR cuts the attack gain by 81% and raises cold-start Recall@100 from 0.1133 to 0.170, while keeping the merged NDCG@10 within 4% of the undefended system.

Figures & tables

Appendix figures & tables4 assets

Supplementary material from the paper’s appendix.

Appendix

Explore similar work

CardsList
  1. Compressed Video Aggregator: Content-driven Module for Efficient Micro-Video Recommendation

    May 9, 2026Yang Xiao, Huiyuan Chen, Kaiyuan Deng +5Video Representation LearningEmbedding Compression

  2. Video-Index: A Curated Meta-Benchmark for Video Understanding

    Oct 1, 2026Enxin Song, Yinuo Xu, Shusheng Yang +2Video UnderstandingVideo-Language Model Evaluation