Daniel Keim and Menna El-Assady of Univ of Konstanz offer an interactive exploration tool for a thoughtfully indexed set of 71 papers on Human-Centered Evaluations of Human-Centered Machine Learning …an-centered-evaluations-star.dbvis.de

Happening now at #eurovis2022, Rita Sevastjanova presenting our work on Language Model Explainability (LingVis.io). Check out lmfingerprints.lingvis.io for more information. #NLProc

Happening now at #eurovis2022, I. Baltazar Pérez presenting our typography of guidance tasks in mixed-initiative visual analytics, a collaboration with @dadolissimo @SilviaMiksch13 and Fabian Sperrle. #eurovis

Do you know about the Rationalization Trap in #XAI? We present a conceptual model on challenges that can occur when language model explanations differ from our mental model of language.✍️Menna📢IJCAIconf #IJCAI2022Workshop Read more about it here: arxiv.org/abs/2207.06897

What a cool summary of our recent paper on challenges that may occur when designing language model explanations! 😍 Thanks, @NLPParrot ! 😀 Paper: arxiv.org/abs/2207.06897 Menna Data Analysis and Visualization

To "understand language", masked #languagemodels should capture semantic constraints of function words (e.g., negation, coordination, quantifiers). We show that BERT, RoBERTa, and ALBERT don't: arxiv.org/abs/2209.07836 More: function-words.lingvis.io To appear COLING 2022 . Data Analysis and Visualization

We prepared all our research material as a Linktree to provide a concise overview of all our recent projects. 🚀🚀🚀Enjoy🚀🚀🚀 linktr.ee/visualmusicolo…

We are presenting our work on #VisualMusicology at the AI+X Summit 2022 with ETH AI Center today! Drop by our booth R2 and check it out!

Happening now at #IEEEVIS 2022: Max Fischer presenting live our research on communication analysis with ETH AI Center. Frederik Dennig, Daniel Seebacher, D. Keim Data Analysis and Visualization, and Menna El-Assady IEEE VIS #VDS Drop by or visit the conference website to check it out! communication-analysis.dbvis.de

Our IEEE VIS paper shows how to visually compare embedding spaces produced by adapted language models. Thanks, AdapterHub, for the variety of adapters for evaluation!🥳And a big thank you to my co-authors E.Cakmak Shauli Ravfogel @ryandcotterell Menna El-Assady!😀 adapters.demo.lingvis.io

Today, I'm happy to present "Semantic Color Mapping: A Pipeline for Assigning Meaningful Colors to Text" at the VisGuides workshop at IEEE VIS2022 #VIS22 IEEE VIS. This is joint work with Menna El-Assady Rita Sevastjanova Thilo Spinner, Rebecca Kehlbeck, Udo Schlegel and Fabian Sperrle Data Analysis and Visualization

How should we compare the embedding spaces of large language models? Did language model adapters (AdapterHub) introduce bias? Join us in the current #ieeevis session on comparison in Oklahoma Station 1. #vis22

Nach dem erfolgreichen Kick-Off im April '22 wurde das #UniKonstanz Centre for HUMAN | DATA | SOCIETY gegründet. Das Zentrum untersucht transdisziplinär Aspekte der Datengesellschaft sowie die Interaktionen zwischen Menschen + datengetriebenen Systemen. t1p.de/4lrpm

Thank you to #EuroVis and the Early Career Award Committee for the award. I'm honored and humbled to be working in, and recognized by the amazing #ieeevis community. Looking forward to many more years of #visualization research and future collaborations. ☺️ #eurovis2023

Our templates for video bulletin submissions are now online (vis4pandemres.github.io/cfp/). We also thank the preliminary AP & IPC for their support (see vis4pandemres.github.io/organizers/). Submissions are due July 1st! IEEE VIS #ieeevis

Last paper session at #EuroVis this morning is on visualization and machine learning. Yannick Metz will be presenting our work on interactive state sequence exploration for #Reinforcement_Learning (diglib.eg.org/handle/10.1111…) #eurovis2023 #ieeevis #visualization #visxai

Deadline extension! Workshop submissions are now due July 8th! Find more information here: vis4pandemres.github.io/cfp/ Looking forward to your contributions!

What biases are encoded in texts generated by #LLMs? Our workspace helps to explore stereotypes encoded in prompt outputs and detect unexpected word associations. Menna El-Assady will present our work VDS at KDD and VIS🥳 Paper and demo: prompt-comparison.lingvis.io #ieeevis2023 IEEE VIS

Our visualizations show that larger models, e.g., ChatGPT and Bard, are less biased than their smaller counterparts. Are the models getting better or the engineering steps used to adapt the outputs? 🤔Menna El-Assady #ieeevis #vds 👉prompt-comparison.lingvis.io

Want to explore #wordembedding properties? Here are some ideas for data preparation steps✂️, interesting additional features🍩, and some relevant references✍️: embedding-framework.lingvis.io With Menna El-Assady 😊 #nlviz #ieeevis #LLMs