From b615b7f3e507cdb11f7bde4d0380fe0f483025cb Mon Sep 17 00:00:00 2001
From: Sifat Anindho <102836340+sifatanindho@users.noreply.github.com>
Date: Tue, 12 Nov 2024 17:13:14 -0700
Subject: [PATCH] Update readme.md

I think you guys misspelled `devtest_rw.jsonl` as `devtest_mt.jsonl` in the description for the real-world distribution split. Not a big issue, just something I noticed.
---
 .../controller_data/datasets/NewsMTSC-dataset/readme.md         | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/NewsSentiment/controller_data/datasets/NewsMTSC-dataset/readme.md b/NewsSentiment/controller_data/datasets/NewsMTSC-dataset/readme.md
index afba64f..c5ff2ad 100644
--- a/NewsSentiment/controller_data/datasets/NewsMTSC-dataset/readme.md
+++ b/NewsSentiment/controller_data/datasets/NewsMTSC-dataset/readme.md
@@ -9,7 +9,7 @@ The dataset consists of three splits. In practical terms, we suggest to use the
 
 * `train.jsonl` - For **training**.
 * `devtest_mt.jsonl` - To evaluate a model's classification performance only on sentences that contain **at least two target mentions**. Note that the mentions were extracted to refer to different persons but in a few cases might indeed refer to the same person since we extracted them automatically.
-* `devtest_mt.jsonl` - To evaluate a model's classification performance on a "**real-world**" set of sentences, i.e., the set was created with the objective to resemble real-world distribution as to sentiment and other factors mentioned in the paper.
+* `devtest_rw.jsonl` - To evaluate a model's classification performance on a "**real-world**" set of sentences, i.e., the set was created with the objective to resemble real-world distribution as to sentiment and other factors mentioned in the paper.
 
 
 ### Format