Skip to content

Commit

Permalink
updated tag on franken paper
Browse files Browse the repository at this point in the history
  • Loading branch information
tobiasgerstenberg committed Nov 28, 2023
1 parent aef3dba commit 12e1207
Show file tree
Hide file tree
Showing 10 changed files with 413 additions and 12 deletions.
2 changes: 1 addition & 1 deletion content/publication/franken2023social.md
Original file line number Diff line number Diff line change
Expand Up @@ -13,7 +13,7 @@ publication_types = ["3"]
publication_short = "_Socially Responsible Language Modelling Research Workshop (NeurIPS 2023)_"
publication = "Fränken J., Kwok S., Ye P., Gandhi K., Arumugam D., Moore J., Tamkin A., Gerstenberg T., Goodman N. D. (2023). Social Contract AI: Aligning AI Assistants with Implicit Group Norms. In _Socially Responsible Language Modelling Research Workshop (NeurIPS 2023)."
abstract = "We explore the idea of aligning an AI assistant by inverting a model of users' (unknown) preferences from observed interactions. To validate our proposal, we run proof-of-concept simulations in the economic ultimatum game, formalizing user preferences as policies that guide the actions of simulated players. We find that the AI assistant accurately aligns its behavior to match standard policies from the economic literature (e.g., selfish, altruistic). However, the assistant's learned policies lack robustness and exhibit limited generalization in an out-of-distribution setting when confronted with a currency (e.g., grams of medicine) that was not included in the assistant's training distribution. Additionally, we find that when there is inconsistency in the relationship between language use and an unknown policy (e.g., an altruistic policy combined with rude language), the assistant's learning of the policy is slowed. Overall, our preliminary results suggest that developing simulation frameworks in which AI assistants need to infer preferences from diverse users can provide a valuable approach for studying practical alignment questions."
tags = ["Spotlight at NeurIPS Workshop"]
tags = ["Oral at NeurIPS Workshop"]
image_preview = ""
selected = false
projects = []
Expand Down
2 changes: 1 addition & 1 deletion docs/member/tobias_gerstenberg/index.html
Original file line number Diff line number Diff line change
Expand Up @@ -454,7 +454,7 @@ <h2 id="publications">Publications</h2>



<br><span style="color:blue;"><b>Spotlight at NeurIPS Workshop</b></span>
<br><span style="color:blue;"><b>Oral at NeurIPS Workshop</b></span>


<p>
Expand Down
2 changes: 1 addition & 1 deletion docs/publication/franken2023social/index.html
Original file line number Diff line number Diff line change
Expand Up @@ -261,7 +261,7 @@ <h3>Abstract</h3>
<div class="article-tags">
<p>

<span style="color:blue;"><b>Spotlight at NeurIPS Workshop</b></span><br>
<span style="color:blue;"><b>Oral at NeurIPS Workshop</b></span><br>

</p>
</div>
Expand Down
2 changes: 1 addition & 1 deletion docs/publication/index.html
Original file line number Diff line number Diff line change
Expand Up @@ -1674,7 +1674,7 @@ <h1>Publications</h1>



<br><span style="color:blue;"><b>Spotlight at NeurIPS Workshop</b></span>
<br><span style="color:blue;"><b>Oral at NeurIPS Workshop</b></span>


<p>
Expand Down
8 changes: 4 additions & 4 deletions docs/sitemap.xml
Original file line number Diff line number Diff line change
Expand Up @@ -37,7 +37,7 @@

<url>

<loc>https://cicl.stanford.edu/publication_types/</loc>
<loc>https://cicl.stanford.edu/tags/oral-at-neurips-workshop/</loc>
<lastmod>2023-10-30T00:00:00+00:00</lastmod>
<priority>0</priority>
</url>
Expand All @@ -46,17 +46,17 @@

<url>

<loc>https://cicl.stanford.edu/publication/franken2023social/</loc>
<loc>https://cicl.stanford.edu/publication_types/</loc>
<lastmod>2023-10-30T00:00:00+00:00</lastmod>
<priority>0</priority>
</url>



<url>

<loc>https://cicl.stanford.edu/tags/spotlight-at-neurips-workshop/</loc>
<loc>https://cicl.stanford.edu/publication/franken2023social/</loc>
<lastmod>2023-10-30T00:00:00+00:00</lastmod>
<priority>0</priority>
</url>


Expand Down
2 changes: 1 addition & 1 deletion docs/tags/index.html
Original file line number Diff line number Diff line change
Expand Up @@ -239,7 +239,7 @@ <h1 class="pt-3">Tags</h1>


<div>
<h2><a href="https://cicl.stanford.edu/tags/spotlight-at-neurips-workshop/">Spotlight at NeurIPS Workshop</a></h2>
<h2><a href="https://cicl.stanford.edu/tags/oral-at-neurips-workshop/">Oral at NeurIPS Workshop</a></h2>
<div class="article-style">

</div>
Expand Down
6 changes: 3 additions & 3 deletions docs/tags/index.xml
Original file line number Diff line number Diff line change
Expand Up @@ -13,11 +13,11 @@


<item>
<title>Spotlight at NeurIPS Workshop</title>
<link>https://cicl.stanford.edu/tags/spotlight-at-neurips-workshop/</link>
<title>Oral at NeurIPS Workshop</title>
<link>https://cicl.stanford.edu/tags/oral-at-neurips-workshop/</link>
<pubDate>Mon, 30 Oct 2023 00:00:00 +0000</pubDate>

<guid>https://cicl.stanford.edu/tags/spotlight-at-neurips-workshop/</guid>
<guid>https://cicl.stanford.edu/tags/oral-at-neurips-workshop/</guid>
<description></description>
</item>

Expand Down
Loading

0 comments on commit 12e1207

Please sign in to comment.