IMC: A Benchmark for Invariant Learning under Multiple Causes
Published in CVPR 2025 Workshop on Domain Generalization: Evolution, Breakthroughs and Future Horizon, 2025
We introduce IMC, a benchmark that exposes fundamental failure modes of existing domain generalization methods under multiple causally-intertwined spurious correlations—a realistic but underexplored setting that current benchmarks fail to capture.
Award: Oral Presentation & Best Paper Award (Cash Prize: USD 750)