Towards Debiasing Frame Length Bias in Text-Video Retrieval via Causal Intervention

Published in BMVC 2023, 2023

Burak Satar, Hongyuan Zhu, Hanwang Zhang, Joo-Hwee Lim

Shows that text-video retrieval models exploit clip length as a shortcut, and mitigates the bias with causal intervention.

PDF (BMVC) arXiv Project Page Video

Towards Debiasing Frame Length Bias in Text-Video Retrieval via Causal Intervention figure

BibTeX
@inproceedings{DBLP:conf/bmvc/SatarZZL23,
  author       = {Burak Satar and
                  Hongyuan Zhu and
                  Hanwang Zhang and
                  Joo{-}Hwee Lim},
  title        = {Towards Debiasing Frame Length Bias in Text-Video Retrieval via Causal
                  Intervention},
  booktitle    = {34th British Machine Vision Conference 2023, {BMVC} 2023, Aberdeen,
                  UK, November 20-24, 2023},
  pages        = {650--658},
  publisher    = {{BMVA} Press},
  year         = {2023},
  url          = {http://proceedings.bmvc2023.org/650/},
  timestamp    = {Wed, 15 Oct 2025 07:23:59 +0200},
  biburl       = {https://dblp.org/rec/conf/bmvc/SatarZZL23.bib},
  bibsource    = {dblp computer science bibliography, https://dblp.org}
}