[visionlist] Artificial Social Intelligence Workshop and Challenge @ ICCV 2023 - Call for Papers

Louis-Philippe Morency morency at cs.cmu.edu
Mon Jun 12 20:50:58 -04 2023

Hello!    Bonjour!

We welcome and invite you to participate in the ICCV 2023 Challenge and
Workshop on “Artificial Social Intelligence” which will be held on October
2, 2023 in Paris at ICCV.


*Event: *Challenge and Workshop Artificial Social Intelligence

*Location: *Co-located with ICCV 2023

*Date: *2 October 2023

*Paper Submission Deadline: *21 July 2023

*Workshop Website: *https://sites.google.com/view/asi-iccv-2023

*Social-IQ Challenge Website: *


*Workshop Overview: Artificial Social Intelligence*

As social agents, humans have *social intelligence* that enables them to
engage with others through complex signals such as facial expressions, body
motion, and speech, among other communication modalities. Given the
monumental utility of social intelligence in the daily lives of humans,
there has been an emerging interest in building social intelligence in AI
systems. We welcome researchers across disciplines (computer vision, NLP,
robotics, neuroscience, etc) to join our workshop and discuss the science
of social intelligence, as well as underlying technical challenges, goals,
and ethics of building social intelligence in real-world AI systems. Our
workshop will include keynote talks, panel discussions, and paper

*A new Social-IQ Challenge*

Our workshop will host the Social-IQ challenge from CMU’s MultiComp Lab on
multimodal Question-Answering for socially-intelligent video understanding.
Our Social-IQ benchmark contains over 1,000 videos, over 6,000 questions,
and over 24,000 answers. There are $1,200 in total prizes for challenge
participants. In order to encourage diverse, innovative approaches
addressing the Social-IQ 2.0 challenge, we welcome submissions under one of
the following three focuses:

   - *few-shot focus*: composing pre-trained models in a zero- or few-shot
   - *fusion and reasoning *focus: learning to fuse and integrate
   information from multimodal features
   - *representation learning focus*: Learning new multimodal
   representations directly from raw data

Additional details regarding challenge participation can be found on our
challenge website:


*Call for Papers and Participation*

There will be 2 paper tracks for this workshop, both included in ICCV
proceedings :

   - Social-IQ Challenge track: where participants of the Social-IQ
   challenge summarize their findings and approach.
   - Workshop track: for all researchers interested in topics related to
   artificial social intelligence, with no need to use the Social-IQ dataset

General workshop papers can be focused on any topics related to artificial
social intelligence, including (but not limited to) the following areas:

   - Social signal processing (e.g., turn-taking prediction)
   - Social behavior, emotion, pose, and/or trajectory prediction by
   virtual and embodied agents
   - Social behavior and emotion generation in virtual and embodied agents
   - Applications of artificial social intelligence

*Important Dates*

   - Paper submission deadline:        21 July 2023
   - Notification of acceptance:         4 August 2023
   - Camera-ready deadline:               11 August 2023
   - Workshop day:                             2 October 2023

*Organizing Committee*

   - Evonne Ng, UC Berkeley
   - Alex Wilf, Carnegie Mellon University
   - Leena Mathur, Carnegie Mellon University
   - Paul Pu Liang, Carnegie Mellon University
   - Trevor Darrell, UC Berkeley
   - Hanbyul Joo, Seoul National University
   - Angjoo Kanazawa, UC Berkeley
   - Hyun Soo Park, University of Minnesota
   - Shiry Ginosar, UC Berkeley
   - Justine Cassell, Carnegie Mellon University, Inria Paris
   - Louis-Philippe Morency, Carnegie Mellon University

*Contact *

If you have any questions about the workshop, please email Alex Wilf (
awilf at cs.cmu.edu), Leena Mathur (lmathur at cs.cmu.edu), or Evonne Ng (
evonne_ng at berkeley.edu)
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://visionscience.com/pipermail/visionlist_visionscience.com/attachments/20230612/147e46c6/attachment-0001.html>

More information about the visionlist mailing list