Update README.md

README.md

@@ -31,6 +31,10 @@ This makes it significantly more scalable than our previous approaches.
 * _[Snorkel: Rapid Training Data Creation with Weak Supervision](https://arxiv.org/abs/1711.10160) [VLDB 2018]_
 * _[Data Programming: Creating Large Training Sets, Quickly](https://arxiv.org/abs/1605.07723) [NIPS 2016]_
 
+## Q&A
+One of the best places to look for Q&A (besides the doc strings) is the Issues page.
+We tag issues that might be aprticularly helpful with the "reference question" label; see them [here](https://github.com/HazyResearch/metal/issues?utf8=%E2%9C%93&q=is%3Aissue+label%3A%22reference+question%22).
+
 ## Sample Usage
 This sample is for a single-task problem. 
 For a multi-task example, see tutorials/Multitask.ipynb.