Joseph SalmonOct 31, 2023
My student Tanguy Lefort is going to #HCOMP2023 (at TU Delft, Nov. 6-10).
For those interested in #crowdsourcing and ambiguous tasks detection - or other things ! - feel free to come and have a chat with him if you are there.
A recap below of his PhD so far and he has been working on since 2021:
WikiResearchSep 7, 2023
RT by @wikiresearch: 📣 #HCOMP2023 is hosting our Workshop on Knowledge Integrity in Wikipedia and Collaborative Projects.
https://www.knowledgeintegrity.wiki/ https://twitter.com/julianvicens/status/1699449848821436528
wikiKI: Knowledge Integrity on Wikipedia and Collaborative Projects – workshop at HCOMP 2023

Knowledge Integrity on Wikipedia and Collaborative Projects - workshop at HCOMP 2023

Amy ZhangAug 25, 2023
New paper #hcomp2023 led by an ugrad in my lab, Andre Ye and
@cqz! We look at how to represent uncertainty in 2D annotations for computer vision models. In this case, we focus on medical image segmentation, where uncertainty is important for experts to interpret. https://arxiv.org/abs/2308.07528
Confidence Contours: Uncertainty-Aware Annotation for Medical Semantic Segmentation

Medical image segmentation modeling is a high-stakes task where understanding of uncertainty is crucial for addressing visual ambiguity. Prior work has developed segmentation models utilizing probabilistic or generative mechanisms to infer uncertainty from labels where annotators draw a singular boundary. However, as these annotations cannot represent an individual annotator's uncertainty, models trained on them produce uncertainty maps that are difficult to interpret. We propose a novel segmentation representation, Confidence Contours, which uses high- and low-confidence ``contours'' to capture uncertainty directly, and develop a novel annotation system for collecting contours. We conduct an evaluation on the Lung Image Dataset Consortium (LIDC) and a synthetic dataset. From an annotation study with 30 participants, results show that Confidence Contours provide high representative capacity without considerably higher annotator effort. We also find that general-purpose segmentation models can learn Confidence Contours at the same performance level as standard singular annotations. Finally, from interviews with 5 medical experts, we find that Confidence Contour maps are more interpretable than Bayesian maps due to representation of structural uncertainty.

arXiv.org
Jeffrey P. Bigham 🔥🔥May 7, 2023
Whose brilliant idea was it to AC for #uist2023, #assets2023, #hcomp2023 and SC for #emnlp2023 all within about 6 weeks? https://media.giphy.com/media/3o6nVb3n424wXYDUJi/giphy.gif?cid=ecf05e470stct6mkxosxtsdpvrxh3cg6zq1r759r4wqpo7ce&ep=v1_gifs_search&rid=giphy.gif&ct=g
- Find & Share on GIPHY

The official GIPHY channel for Team Coco. Watch CONAN weeknights @ 11/10c on TBS Network. #CONAN

GIPHY