We are glad to announce that 1 paper authored by our group has been accepted at CLIC-it 2023:
Two papers from our lab have been accepted to #eacl2027 🎉
1⃣ RedVox: Safety and Fairness Gaps in Speech Models Across Languages (Main) by @BeatriceSavoldi, @sarapapi, @aissawafa94, @negri_teo, @luisabentivogli
🖨️ Paper: https://arxiv.org/pdf/2606.26968
🤗 Data: https://huggingface.co/datasets/FBK-MT/RedVox
Meet the new PhD additions to our group! 🧑💻
Andrea DeCarlo he'll be working on multimedia personalization with multimodal large language models
And Harun Eroglu, he'll be working on human-centred evaluation frameworks for multilingual technologies.
#MT #NLP #AI #FBK
Our #PickOfTheWeek 📚: “Training Dynamics Impact Post-Training Quantization Robustness” by @actatjer, @n_ajroldi, @jonasgeiping
Pick of the week @fbk_mt
Cool paper on how robustness to quantization in LLMs is also impacted by training conditions, instead of just the number of training tokens!
https://arxiv.org/abs/2510.06213