https://openreview.net/forum?id=b0qRSUcQP7&referrer=%5Bthe%20profile%20of%20Xianpei%20Han%5D(%2Fprofile%3Fid%3D~Xianpei_Han1)
Multimodal Reward Models (MM-RMs) are crucial for aligning Large Language Models (LLMs) with human preferences, particularly as LLMs increasingly interact with...
the devilspurious correlationsdetailstacklingunimodal
https://www.wnycstudios.org/podcasts/otm/articles/spurious-correlations
It is by now an age old adage that "correlation doesn't equal causation," but the internet just loves stories that make spurious correlations. Just...
swimming poolnicholas cagedeathsdrowningvs
https://openreview.net/forum?id=w97lDmoD0U&referrer=%5Bthe%20profile%20of%20Marzyeh%20Ghassemi%5D(%2Fprofile%3Fid%3D~Marzyeh_Ghassemi2)
Benchmarks for out-of-distribution (OOD) generalization often reveal a strong positive correlation between in-distribution (ID) and OOD accuracy across models,...
out ofspurious correlationsaggregationhidesdistribution
https://openreview.net/forum?id=f6GVzoTLez&referrer=%5Bthe%20profile%20of%20Raghav%20Mehta%5D(%2Fprofile%3Fid%3D~Raghav_Mehta1)
Deep learning models can perform well in complex medical imaging classification tasks, even when basing their conclusions on spurious correlations (i.e....
in thespurious correlationsdebiasingcounterfactualspresence
https://kottke.org/14/05/spurious-correlations?utm_source=nextdraft&utm_medium=website
Tyler Vigen is collecting examples of data that correlate closely but are (probably) otherwise unrelated. Remember kids, correlation != causation.
spurious correlations