ELOQUENCEAI

🤖 Friday Fact: AUC – Area Under the ROC Curve

Fact: AUC (Area Under the ROC Curve) measures how well a classifier can distinguish between positive and negative classes across all possible threshold settings. Its value ranges from 0 to 1.

#ELOQUENCE #AI #FridayFact #DataScience

🌟 Join us for the next ELOQUENCE Webcafé with Telefónica! 🌟

🗓 Date: 15 October 2025
⏰ Time: 10:00 (CEST)
📍 Platform: Zoom

📌 Save your spot now: us02web.zoom.us/meeting/regi...

Stay tuned – more details about our speakers are coming soon!
#ELOQUENCE #AI #Webcafé #SmartHomes

YouTube video by ELOQUENCE AI ELOQUENCE Explainer #4 | Explainability in AI: Making Dialogue Systems Transparent

📌Explainability in AI: Making Dialogue Systems Transparent

🎤In this video, Dairazalia Sanchez-Cortes from IDIAP Research Institute, explores the concept of explainability in artificial intelligence.

🔗Find out more in the video: youtu.be/R8bsnOpSsmM

#ELOQUENCEAI #ExplainerVideo #LLMs

🍂Kicking Off Autumn with Fresh Updates

📬 Check out our latest ELOQUENCE newsletter!

🔗Read the full newsletter at: eloquenceai.eu/publications/

#ELOQUENCEAI #Newsletter #LLMs #TrustworthyAI #MultilingualAI

🚀 Friday AI Fact

A ROC Curve (Receiver Operating Characteristic Curve) is a graphical plot that illustrates the diagnostic ability of a binary classifier system.

#ELOQUENCE #FridayFact #AI #MachineLearning #DataScience #AIInsights #ModelEvaluation

🚀 Friday AI Fact: F1 Score

Precision tells us how many predicted positives are actually positive, while recall shows how many actual positives we successfully identified. The F1 score combines these into a single number, giving a balanced view of a model’s performance.

#ELOQUENCE

🌟 Meet Anca Radu, AI & human rights expert! 🌟

Discover her work on legal & ethical AI in the judiciary, her role at CAIDP Europe and more.

👩‍💻 Learn more: eloquenceai.eu/experts

#ELOQUENCEAI #AIethics #CommunityOfExperts

YouTube video by ELOQUENCE AI ELOQUENCE | Virtual demo days | PILOT 3

✨ Virtual demo days | PILOT 3

🔊 In this pilot, we aim to assess the effectiveness of our proposed LLM-based virtual agent within the demanding customer service sector.

🔗Find out more in the video: www.youtube.com/watch?v=iqUQ...

#ELOQUENCEAI #ConversationalAI #MultilingualAI

We are thrilled to announce the upcoming European Parliament’s Multilingualism Day 2025, a unique celebration of linguistic diversity in the EU!

✨Online workshops → Friday, 26 September (registrations are open and you can register here:
ep-events.europarl.europa.eu/europarl/MLD...

#ELOQUENCEAI

🧠 Friday AI Fact: What is Accuracy in AI?

🎯Accuracy is one of the most common metrics used to evaluate a model's performance.

It represents the ratio of correctly predicted instances to the total number of predictions made.

Same time next Friday – new AI fact coming your way!

#ELOQUENCEAI

📢 New ELOQUENCE Publication Alert!

Our partners released a new study: “Scaling and Enhancing LLM-based AVSR: A Sparse Mixture of Projectors Approach”.

🔗 Read the full paper here: eloquenceai.eu/publications/

#ELOQUENCE #AI

YouTube video by ELOQUENCE AI ELOQUENCE | Virtual demo days | PILOT 2

✨ Virtual demo days | PILOT 2

🔊 This pilot leverages open-source data to craft scenarios and benchmarks that test language models against social norms and detect biases in sentence construction.

🔗Find out more in the video: www.youtube.com/watch?v=sahj...

#ELOQUENCEAI #ConversationalAI

🧠 Friday AI Fact: What is Recall?

Recall (or sensitivity) is the ratio of correctly predicted positive observations to all observations in the actual class.

🚀 Catch you next Friday with another AI fact!

#ELOQUENCE #AI #MachineLearning #Recall #FridayAIFact #ModelMetrics

YouTube video by ELOQUENCE AI Detecting and Mitigating Bias in AI Systems: Webcafé with Javi García Gilabert and Miguel Claramunt

✨ For everyone who couldn’t join our Webcafé session last Friday – the recording is now online!

▶️ Watch the full session here: youtu.be/AHfP29vU6k4

🔗Learn more about MT-Lens here: eloquenceai.eu/publications/

🌞 Stay tuned for more exciting content from the ELOQUENCE project!

#ELOQUENCE #AI

🚨 ELOQUENCE Webcafé starts in less than 3 hours!

⏳ It’s not too late to secure your spot: us02web.zoom.us/meeting/regi...

📩 If you’ve already registered, don’t forget to check your confirmation email for instructions on how to join the session.

See you soon at the Webcafé! ☕✨

#ELOQUENCE

📢 Reminder: Our next ELOQUENCE Webcafé is coming up this Friday!

🎙️ Speakers: Miguel Claramunt and Javi García Gilabert
🎯 Topic: Detecting and Mitigating Bias in AI Systems
📅 Date: Friday, 22 August 2025
⏰ Time: 12:00 PM (CEST)

💡 Save your spot now at: us02web.zoom.us/meeting/regi...

☕ Join us this Friday for a new ELOQUENCE Webcafé session!

We’ll discuss Detecting and Mitigating Bias in AI Systems with Javier García Gilabert and Miguel Claramunt from the Barcelona Supercomputing Center.

📌 Save your spot at: us02web.zoom.us/meeting/regi...

#ELOQUENCE

📊Friday AI Fact: What is a Confusion Matrix?

A confusion matrix is a table that helps you understand how well your classification model is performing.

Same time next Friday - new AI fact coming your way!

#ELOQUENCE #AI #MachineLearning #ConfusionMatrix

YouTube video by ELOQUENCE AI ELOQUENCE | Virtual demo days | PILOT 1

✨Virtual demo days – PILOT 1 ✨

Discover the first steps toward building trustworthy multilingual conversational AI!

This video captures the presentation of Pilot 1, shared during the ELOQUENCE Plenary Meeting.

🔗Find out more: www.youtube.com/watch?v=QU4G...

#ELOQUENCEAI

YouTube video by ELOQUENCE AI ELOQUENCE Explainer #3 | From Audio to Insight: Teaching LLMs to Comprehend Speech

✨ In this short explainer, Oldřich Plchot, from Brno University of Technology, introduces the motivation and methods behind a novel approach to enabling Large Language Models to understand human speech in a meaningful, grounded way.

🔗Find out more: www.youtube.com/watch?v=xz4Y...

#ELOQUENCEAI

🧠 Friday AI Fact: What is Validation Data?

📊 A validation dataset is used to provide an unbiased evaluation of a model during the training phase.

🤖 Don’t miss next Friday’s AI fact!

#ELOQUENCE #AI #MachineLearning #FridayAIFact #DataScience #AIExplained #ResponsibleAI

🌟 Community of Experts Series: Giuseppe Caggianese 🌟

Giuseppe Caggianese is a Research Scientist at the National Research Council of Italy (CNR), Institute for High-Performance Computing and Networking (ICAR-CNR).

🔗 Find out more at: eloquenceai.eu/experts

#ELOQUENCEAI #CommunityOfExperts

🧠 Friday AI Fact: Training Data! 📚🤖

Training data in AI is a dataset used to teach a model how to make predictions or recognize patterns.

Catch you next Friday with another AI fact!

#ELOQUENCEAI #FridayFacts #TrainingData #MachineLearning #AIexplained

🌟 Community of Experts Series: Dr Alberto Quintavalla 🌟

Dr Alberto Quintavalla is Assistant Professor at the Erasmus University Rotterdam and affiliated researcher of the Jean Monnet Centre on Digital Governance.

🔗 Meet more experts at: eloquenceai.eu/experts

#ELOQUENCEAI #CommunityOfExperts

🎙️ Facts and Norms in Human Rights Assessment: Can human rights compliance be measured?

🔗 Here you can watch the webinar with Professor Martin Scheinin and explore the intersection of AI and human rights :
hrjust-intersect-observatory.eu/events/facts...

#ELOQUENCEAI #HumanRights

🤖 Friday AI Fact: Hyperparameters!

A hyperparameter in AI is a parameter whose value is set before the learning process begins.

Unlike model parameters, hyperparameters control how the learning happens, like the learning rate, number of layers or batch size.

#ELOQUENCEAI #FridayFacts

✨The Horizon Europe-funded ELOQUENCE project hit its mid-point at a plenary hosted by Telefónica in Barcelona. Bringing together 16 partners, the project advances ethical, multilingual, and bias-free conversational AI.

🔗 Learn more: eloquenceai.eu/eloquence-pr...

#ELOQUENCEAI #HorizonEurope

YouTube video by ELOQUENCE AI ELOQUENCE Explainer #2 | Ethical AI and Human Rights - Shaping Trustworthy Technologies

🧠 Ethical AI and Human Rights: Shaping Trustworthy Technologies

💡 In this explainer video, we dive with partners from European University Institute, this time esteemed Dr Martin Scheinin.

🔗 Learn more at: www.youtube.com/watch?v=T3Z1...

#ELOQUENCEAI #EthicalAI #MultilingualAI

🧠 Friday AI Fact: Variance! 🔍📊

Variance measures how much the values in a dataset differ from the mean. In simple terms it tells us how spread out the data is.

Catch you next Friday for another AI fact!

#ELOQUENCEAI #FridayFacts #Variance #AIexplained

🌟 Community of Experts Series: Theofanis Orphanoudakis 🌟

Theofanis Orphanoudakis is an associate professor at HOU and director of the DSMC Lab as well as a research associate and independent consultant.

🔗Read more at: eloquenceai.eu/experts

#ELOQUENCEAI #CommunityOfExperts