diff --git a/docs/src/concepts/flux_factors.md b/docs/src/concepts/flux_factors.md
index cab6bb315..7fc02aea5 100644
--- a/docs/src/concepts/flux_factors.md
+++ b/docs/src/concepts/flux_factors.md
@@ -1,4 +1,4 @@
 # Incorporating Neural Network Factors
 
-IncrementalInference.jl and RoME.jl has native support for using Neural Networks (via [Flux.jl](https://fluxml.ai/Flux.jl/stable/)) as non-Gaussian factors.  Documentation is forthcoming, but meanwhile [see the following generic Flux.jl factor structure](https://github.com/JuliaRobotics/IncrementalInference.jl/tree/master/src/Flux).  Note also that a standard [`Mixture` approach already exists too](https://github.com/JuliaRobotics/RoME.jl/blob/master/src/factors/flux/MixtureFluxPose2Pose2.jl).
+IncrementalInference.jl and RoME.jl has native support for using Neural Networks (via [Flux.jl](https://fluxml.ai/Flux.jl/stable/)) as non-Gaussian factors.  Documentation is forthcoming, but meanwhile [see the following generic Flux.jl factor structure](https://github.com/JuliaRobotics/IncrementalInference.jl/blob/master/ext/IncrInfrFluxFactorsExt.jl).  Note also that a standard [`Mixture` approach already exists too](https://github.com/JuliaRobotics/RoME.jl/blob/master/ext/factors/MixtureFluxPose2Pose2.jl).