diff --git a/error-prone-contrib/README.md b/error-prone-contrib/README.md
index a9eddf26576..db63ff439dc 100644
--- a/error-prone-contrib/README.md
+++ b/error-prone-contrib/README.md
@@ -97,7 +97,7 @@ The following is a list of checks we'd like to see implemented:
   `java.util.stream.Collectors` and the various Guava collector factory
   methods.
 - A check which replaces `new Random().someMethod()` calls with
-  `ThreadLocalRandom.current().someMethod()` calls, so as to avoid unnecessary
+  `ThreadLocalRandom.current().someMethod()` calls, to avoid unnecessary
   synchronization.
 - A check which drops `this.` from `this.someMethod()` calls and which
   optionally does the same for fields, if no ambiguity arises.
@@ -214,7 +214,7 @@ Refaster's expressiveness:
   pure expression. Introduce a way to express such a constraint. For example,
   rewriting `optional1.map(Optional::of).orElse(optional2)` to
   `optional1.or(() -> optional2)` is not behavior preserving if evaluation of
-  `optional2` has side-effects.
+  `optional2` has side effects.
 - Similarly, certain refactoring operations are only valid if one of the
   matched expressions is not `@Nullable`. It'd be nice to be able to express
   this.
@@ -241,20 +241,17 @@ Refaster's expressiveness:
   example would be to rewrite `BigDecimal.valueOf("<some-long-value>")` to
   `BigDecimal.valueOf(theParsedLongValue)`.
 - More generally, investigate ways to plug in fully dynamic behavior, e.g.
-  by providing hooks which enable plugging in arbitrary predicates/transformations.
-  The result would be a Refaster/`BugChecker` hybrid. A feature such as this
-  could form the basis for many other features listed here. (As a
-  concrete example, consider the ability to reference
-  `com.google.errorprone.matchers.Matcher` implementations.)
-- Provide a way to match lambda expressions and method references which match a
-  specified functional interface. This would allow rewrites such as
-  `Mono.fromCallable(this::doesNotThrowCheckException)` ->
-  `Mono.fromSupplier(this::doesNotThrowCheckException)`.
-- Provide an extension API which enables defining methods or expressions based on
-  functional properties. A motivating example is the Java Collections
+  by providing hooks which enable plugging in arbitrary predicates/
+  transformations.
+  The result would be a Refaster/`BugChecker` hybrid. A feature like this could
+  form the basis for many other features listed here. (As a concrete example,
+  consider the ability to reference `com.google.errorprone.matchers.Matcher`
+  implementations.)
+- Provide an extension API that enables defining methods or expressions based
+  on functional properties. A motivating example is the Java Collections
   framework, which allows many ways to define (im)mutable (un)ordered
   collections with(out) duplicates. One could then express things like "match
-  any method call which collects its inputs into an immutable ordered list". An
+  any method call that collects its inputs into an immutable ordered list". An
   enum analogous to `java.util.stream.Collector.Characteristics` could be used.
   Out of the box JDK and Guava collection factory methods could be classified,
   with the user having the option to extend the classification.
@@ -262,7 +259,7 @@ Refaster's expressiveness:
   Provide two additional modes: (a) match and drop the comments or (b)
   transport the comments to before/after the replaced expression.
 - Extend Refaster to drop imports that become unnecessary as a result of a
-  refactoring. This e.g. allows one to replace a statically import TestNG
+  refactoring. This e.g. allows one to replace a statically imported TestNG
   `fail(...)` invocation with a statically imported equivalent AssertJ
   `fail(...)` invocation. (Observe that without an import cleanup this
   replacement would cause a compilation error.)
@@ -270,7 +267,7 @@ Refaster's expressiveness:
   methods. For a motivating example see google/error-prone#568.
 - When matching explicit type references, also match super types. For a
   motivating example, see the two subtly different loop definitions in
-  `CollectionRemoveAllFromCollectionBlock`.
+  `CollectionRemoveAllFromCollectionExpression`.
 - Figure out why Refaster sometimes doesn't match the correct generic overload.
   See the `AssertThatIterableHasOneComparableElementEqualTo` template for an
   example.