update

_data/papers/2022.yml

@@ -5,11 +5,11 @@
   selected: y
   year: 2021
   img: hallucinations-min.png
-  title: "Looking for a needle in a haystack: A comprehensive study of hallucinations in neural machine translation"
+  title: "Looking for a Needle in a Haystack: A Comprehensive Study of Hallucinations in Neural Machine Translation"
   authors: "Nuno Guerreiro, <u>Elena Voita</u>,  André Martins"
   doc-url:
-  conf_name:
-  conf_year: 2022
+  conf_name: EACL
+  conf_year: 2023
   url: "https://arxiv.org/pdf/2208.05309.pdf"
   code:
   blog:

index.md

@@ -13,17 +13,26 @@ description: I'm Elena (Lena) Voita and I work on Natural Language Processing. M
 <img src="/img/people/orange_lena-min.jpg" name="readmore_1" width=204px height=240px></a> -->
 
 
-I am Elena (Lena) Voita, a __Research Scientist__ at [__FAIR__](https://ai.facebook.com) working on Natural Language Processing.
+I am Elena (Lena) Voita, a __Research Scientist__ at [__FAIR__](https://ai.facebook.com) working on Analysis and Interpetability for NLP. 
+
+Also, [__NLP Course__ <span style="color:#92bf32">__For You__</span>](https://lena-voita.github.io/nlp_course.html) creator and current [SIGREP](https://www.sigrep.org) president.
+<!--* a __Research Scientist__ at [__FAIR__](https://ai.facebook.com) working on NLP,* current [SIGREP](https://www.sigrep.org) president,* NLP Course <span style="color:#92bf32">For You</span> creator - look [here](https://lena-voita.github.io/nlp_course.html).-->
+
 
 
 Before that, I was Ph.D. student at the University of Edinburgh supervised by [Ivan Titov]({{site:ivan_page}}) and [Rico Sennrich]({{site.rico_page}}),  was awarded [Facebook PhD Fellowship](https://research.fb.com/blog/2020/01/announcing-the-recipients-of-the-2020-facebook-fellowship-awards/),
 worked as a Research Scientist at [Yandex Research]({{site.yandex_research_main}}) side by side with the [Yandex Translate](https://translate.yandex.com) team, 
 while enjoying writing [blog posts](https://lena-voita.github.io/posts.html) and [teaching](https://lena-voita.github.io/nlp_course.html)<a onMouseOver="document.fox.src='/img/people/foxie.jpeg';" onMouseOut="document.fox.src='/img/people/orange_lena-min.jpg';">.</a> 
 
-<span style="color:red">__Blog-course:__</span>    NLP Course <span style="color:#92bf32">For You</span> - look [here](https://lena-voita.github.io/nlp_course.html).
+<!--<span style="color:red">__Blog-course:__</span>    NLP Course <span style="color:#92bf32">For You</span> - look [here](https://lena-voita.github.io/nlp_course.html).-->
 
 ## <span style="color:darkblue">News </span>
 
+__2023__
+
+* __January__ <span style="color:#888"><u>Invited talks:</u></span> [TTI Chicago](https://www.ttic.edu/young-researcher/), [RISE Learning Machines seminar](https://www.ri.se/en/learningmachinesseminars).
+* __Jan-Apr__ <span style="color:#888"><u>Area Chair:</u></span> [ACL 2023](https://2023.aclweb.org).
+
 __2022__
 
 * __Dec__ <span style="color:#888"><u>Keynote:</u></span> [BlackBoxNLP workshop at EMNLP 2022](https://blackboxnlp.github.io).

nlp_course/seq2seq_and_attention.html

@@ -511,8 +511,8 @@ <h2>The Problem of Fixed Encoder Representation</h2>
              style="max-width:60%; margin:20px;float:right;"/>
 
             <p>In the models we looked at so far, the encoder compressed the whole source sentence into a single
-                vector. This can very hard - the number of possible
-                meanings of source is infinite. When the encoder is forced to put all information into
+                vector. This can be very hard - the number of possible
+                source sentences (hence, their meanings) is infinite. When the encoder is forced to put all information into
                 a single vector, it is likely to forget something.
             </p>
             <p><font class="data_text" color="#888"><u>Lena</u>: Imagine

talks.md

@@ -14,6 +14,7 @@ Senior Area Chair:
 Area Chair:
 * EMNLP 2022
 * EACL 2023
+* ACL 2023
 
 Program committee:
 * EMNLP: 2019 <span style="color:#999">(outstanding reviewer)</span>, 2020, 2021
@@ -49,6 +50,14 @@ Program committee:
 
 ### <span style="color:darkblue"> Invited talks </span>
 
+* 11/2022-01/2023 <span class="talk-title">_A Journey on Interpretability Methods in NLP_</span>
+<br/>&nbsp; &nbsp; &nbsp; EPFL
+<br/>&nbsp; &nbsp; &nbsp; [TTI Chicago](https://www.ttic.edu/young-researcher/)
+<br/>&nbsp; &nbsp; &nbsp; [RISE Learning Machines seminar](https://www.ri.se/en/learningmachinesseminars)
+
+* 07/2022 <span class="talk-title">_From Count-Based Methods to GPT-3: Main Developments in Natural Language Processing_</span>
+<br/>&nbsp; &nbsp; &nbsp; [SICSS-Oxford](https://sicss.io/2022/oxford/)
+
 * 10-11/2021 <span class="talk-title">_Neural Machine Translation Inside Out_</span> 
 <br/>&nbsp; &nbsp; &nbsp; [Helsinki LT Seminar](https://blogs.helsinki.fi/language-technology/research-seminar/)
 <br/>&nbsp; &nbsp; &nbsp; ["Shannon meets Turing" Colloquium](https://www.youtube.com/channel/UCjltmKOomhdXQELWi6asoBg)