Update images

_data/authors.yml

@@ -127,6 +127,7 @@ Laurestine Bradford:
   date: Jan 2022
   bio: Language in math and connecting syntax to semantics
   alumni: false
+  avatar: /assets/images/bio/Laurestine Bradford.jpg
   links:
   - label: Website
     url: https://laurestine.github.io/
@@ -135,6 +136,7 @@ Laurestine Bradford:
 Emanuele Bugliarello:
   name: Emanuele Bugliarello
   role: Intern
+  avatar: /assets/images/bio/Emanuele Bugliarello.jpg
   semantic_scholar_id: '83574123'
   alumni: true
   links:
@@ -162,12 +164,14 @@ Xuanda Chen:
   name: Xuanda Chen
   role: PhD
   advisor: Siva Reddy, Timothy J. O'Donnell
+  avatar: /assets/images/bio/Xuanda Chen.jpg
   date: Fall 2022
   bio: Works on language model probing
   alumni: false
   links: []
 Ziling Cheng:
   name: Ziling Cheng
+  avatar: /assets/images/bio/Ziling Cheng.jpg
   role: Intern
   advisor: Jackie CK Cheung
   date: Summer 2022
@@ -202,18 +206,20 @@ Amanda Doucette:
     url: https://github.com/amnda-d
   - label: Scholar
     url: https://scholar.google.com/citations?user=KpJRQbsAAAAJ&hl=en
+
 Nouha Dziri:
   name: Nouha Dziri
   role: Intern
+  avatar: /assets/images/bio/Nouha Dziri.jpg
   semantic_scholar_id: '46217681'
   alumni: true
   links:
   - label: Website
     url: https://webdocs.cs.ualberta.ca/~dziri/
   advisor: Siva Reddy
   date: May 2021
-
   new_role: Postdoctoral young investigator at Allen Institute for AI
+
 Sabina Elkins:
   name: Sabina Elkins
   role: Master
@@ -255,9 +261,16 @@ Lucas Torroba Hennigen:
   role: "Intern"
   advisor: "Siva Reddy"
   new_role: "PhD at MIT Computer Science"
-  avatar: "/assets/images/bio/placeholder/martlet.jpeg"
+  avatar: "/assets/images/bio/Lucas Torroba Hennigen.jpg"
   date: "Fall 2021"
   alumni: true
+  links:
+  - label: "Website"
+    url: "https://ltorroba.github.io/"
+  - label: "GitHub"
+    url: "https://github.com/ltorroba"
+  - label: "Scholar"
+    url: "https://scholar.google.com/citations?user=Zhy1N1sAAAAJ"
 
 Jacob Louis Hoover:
   name: Jacob Louis Hoover
@@ -476,7 +489,10 @@ Neshma Metri:
   bio: Worked on introducing quantifiers in UDepLambda framework which converts dependency
     trees into lambda calculus forms
   alumni: true
-  links: []
+  avatar: /assets/images/bio/Neshma Metri.jpg
+  links:
+  - label: LinkedIn
+    url: https://www.linkedin.com/in/neshma-metri-06b8a1162/
 
 Aristides Milios:
   name: Aristides Milios
@@ -602,8 +618,11 @@ Elinor Poole-Dayan:
   role: "Undergraduate"
   date: "Winter 2021"
   advisor: "Siva Reddy"
-  avatar: "/assets/images/bio/placeholder/martlet.jpeg"
+  avatar: "/assets/images/bio/Elinor Poole-Dayan.jpg"
   alumni: true
+  links:
+  - label: "LinkedIn"
+    url: "https://www.linkedin.com/in/elinor-poole-dayan/"
 
 Ian Porada:
   name: Ian Porada
@@ -613,6 +632,7 @@ Ian Porada:
   date: Winter 2019
   bio: Working on coreference, commonsense knowledge, semantics
   alumni: false
+  avatar: /assets/images/bio/Ian Porada.jpg
   links:
   - label: Website
     url: https://ianporada.github.io/
@@ -637,13 +657,17 @@ Eva Portelance:
     url: https://github.com/evaportelance
   - label: Scholar
     url: https://scholar.google.ca/citations?user=oGYlQJIAAAAJ&hl=en
+
 Victor Prokhorov:
   name: "Victor Prokhorov"
   role: "Intern"
   advisor: "Siva Reddy"
-  avatar: "/assets/images/bio/placeholder/martlet.jpeg"
+  avatar: "/assets/images/bio/Victor Prokhorov.jpg"
   date: "Fall 2021"
   alumni: true
+  links:
+  - label: "Website"
+    url: "https://victorprokhorov.github.io/"
 
 Martin Pömsl:
   name: Martin Pömsl
@@ -689,6 +713,9 @@ Michelle Yang:
   role: "Undergraduate"
   date: "Fall 2021"
   advisor: "Siva Reddy"
-  avatar: "/assets/images/bio/placeholder/martlet.jpeg"
+  avatar: "/assets/images/bio/Michelle Yang.jpg"
   alumni: true
+  links:
+  - label: "Twitter"
+    url: "https://twitter.com/mjieyang"
 

_posts/papers/2022-07-01-nlp_interpretability_survey.md

@@ -5,7 +5,7 @@ categories: Publications # Used to list all posts about publications in /publica
 names: "Andreas Madsen, Siva Reddy, Sarath Chandar" # names of all authors
 link: https://arxiv.org/abs/2108.04840 # link to paper
 twitter: https://twitter.com/andreas_madsen/status/1425794989741727748  # link to twitter thread (optional)
-thumbnail: /assets/images/papers/2021-nlp_interpretability_survey.svg  # link to a thumbnail (optional)
+thumbnail: /assets/images/papers/nlp_interpretability_survey.jpg  # link to a thumbnail (optional)
 venue: ACM Computing Surveys  # venue and year of the paper
 tags: ACM # tag of the paper (exclude year, use shorthand)
 ---

_posts/papers/2023-10-18-MAGNIFICo.md

@@ -12,7 +12,7 @@ tags:
 - Evaluation
 code: https://github.com/McGill-NLP/MAGNIFICo
 webpage: https://mcgill-nlp.github.io/MAGNIFICo
-thumbnail: /assets/images/papers/magnifico.svg
+thumbnail: /assets/images/papers/magnifico.jpg
 categories: Publications
 
 ---

_posts/papers/2024-01-05-2023.findings-emnlp.993.md

@@ -8,7 +8,7 @@ tags:
 link: https://aclanthology.org/2023.findings-emnlp.993/
 code: https://github.com/cesare-spinoso/QCS
 categories: Publications
-
+thumbnail: /assets/images/papers/2023.findings-emnlp.993.jpg
 ---
 
 *{{ page.names }}*
@@ -19,4 +19,4 @@ categories: Publications
 
 ## Abstract
 
-Qualitative coding is a content analysis method in which researchers read through a text corpus and assign descriptive labels or qualitative codes to passages. It is an arduous and manual process which human-computer interaction (HCI) studies have shown could greatly benefit from NLP techniques to assist qualitative coders. Yet, previous attempts at leveraging language technologies have set up qualitative coding as a fully automatable classification problem. In this work, we take a more assistive approach by defining the task of qualitative code suggestion (QCS) in which a ranked list of previously assigned qualitative codes is suggested from an identified passage. In addition to being user-motivated, QCS integrates previously ignored properties of qualitative coding such as the sequence in which passages are annotated, the importance of rare codes and the differences in annotation styles between coders. We investigate the QCS task by releasing the first publicly available qualitative coding dataset, CVDQuoding, consisting of interviews conducted with women at risk of cardiovascular disease. In addition, we conduct a human evaluation which shows that our systems consistently make relevant code suggestions.
+Qualitative coding is a content analysis method in which researchers read through a text corpus and assign descriptive labels or qualitative codes to passages. It is an arduous and manual process which human-computer interaction (HCI) studies have shown could greatly benefit from NLP techniques to assist qualitative coders. Yet, previous attempts at leveraging language technologies have set up qualitative coding as a fully automatable classification problem. In this work, we take a more assistive approach by defining the task of qualitative code suggestion (QCS) in which a ranked list of previously assigned qualitative codes is suggested from an identified passage. In addition to being user-motivated, QCS integrates previously ignored properties of qualitative coding such as the sequence in which passages are annotated, the importance of rare codes and the differences in annotation styles between coders. We investigate the QCS task by releasing the first publicly available qualitative coding dataset, CVDQuoding, consisting of interviews conducted with women at risk of cardiovascular disease. In addition, we conduct a human evaluation which shows that our systems consistently make relevant code suggestions.