diff --git a/R/loo_moment_matching.R b/R/loo_moment_matching.R
index 3721a885..110eff93 100644
--- a/R/loo_moment_matching.R
+++ b/R/loo_moment_matching.R
@@ -26,7 +26,7 @@
 #'   reached, there will be a warning, and increasing `max_iters` may improve
 #'   accuracy.
 #' @param k_threshold Threshold value for Pareto k values above which the moment
-#'   matching algorithm is used. The default value is `1 - 1 / log10(S)`,
+#'   matching algorithm is used. The default value is `min(1 - 1/log10(S), 0.7)`,
 #'   where `S` is the sample size.
 #' @param split Logical; Indicate whether to do the split transformation or not
 #'   at the end of moment matching for each LOO fold.
diff --git a/man/loo_model_weights.Rd b/man/loo_model_weights.Rd
index 7ab44d45..fbcba94c 100644
--- a/man/loo_model_weights.Rd
+++ b/man/loo_model_weights.Rd
@@ -36,7 +36,8 @@ and \eqn{N} is the number of data points. If \code{x} is a list of
 log-likelihood matrices then \code{\link[=loo]{loo()}} is called internally on each matrix.
 Currently the \code{loo_model_weights()} function is not implemented to be used
 with results from K-fold CV, but you can still obtain weights using K-fold
-CV results by calling the \code{stacking_weights()} function directly.}
+CV results by calling the \code{stacking_weights()} or \code{pseudobma_weights()}
+function directly.}
 
 \item{...}{Unused, except for the generic to pass arguments to individual
 methods.}
diff --git a/man/loo_moment_match.Rd b/man/loo_moment_match.Rd
index e0d07ab4..6b1f648d 100644
--- a/man/loo_moment_match.Rd
+++ b/man/loo_moment_match.Rd
@@ -56,7 +56,7 @@ reached, there will be a warning, and increasing \code{max_iters} may improve
 accuracy.}
 
 \item{k_threshold}{Threshold value for Pareto k values above which the moment
-matching algorithm is used. The default value is \code{1 - 1 / log10(S)},
+matching algorithm is used. The default value is \code{min(1 - 1/log10(S), 0.7)},
 where \code{S} is the sample size.}
 
 \item{split}{Logical; Indicate whether to do the split transformation or not