CounterFactual Retrieval for Evidence-Aware Individual Animal Identification

Ankit K. Upadhyay, Aleksandr Algasov, Ekaterina Nepovinnykh, Fedor Zolotarev, Tuomas Eerola et al.|November 4, 2026arXiv

Key Takeaway

Standard retrieval accuracy hides whether models identify animals by their actual features or by background cues; CFR's three-view evaluation exposes this by measuring performance on full images, foreground-only, and background-only versions.

Summary

This paper introduces Counterfactual Foreground Retrieval (CFR), a new evaluation method for animal identification models that reveals whether they rely on actual animal features or background shortcuts.

evaluation multimodal

Key Terms

retrieval evidence-aware-evaluation counterfactual-evaluation