iiinova — where your digital self keeps evolving

Donatella Genovese

@donatellag.bsky.social

73 followers 172 following 7 posts

PhD Student | Works on Explainable AI | prev. visit at LEAP Lab ETH Zurich | donatellagenovese.github.io

Posts Replies Media

Reposted by Donatella Genovese

Pasquale Minervini @neuralnoise.com · 13/03/2025

Please share it within your circles! edin.ac/3DDQK1o

0139

Donatella Genovese @donatellag.bsky.social · 28/02/2025

Really cool paper by @kayoyin.bsky.social about interpretability of In Context Learning, they found that Function Vectors (FV) heads are crucial for few-shot ICL. www.arxiv.org/abs/2502.14010

021

Donatella Genovese @donatellag.bsky.social · 21/02/2025

A really nice resource for understanding how to parallelize LLM training.

020

Donatella Genovese @donatellag.bsky.social · 14/02/2025

🚀 Meta’s new LLM pretraining framework predicts concepts and integrates them into its hidden state to enhance next-token prediction. 🚀 It achieves the same performance with 21.5% fewer tokens and better generalization! 🎯 📝: arxiv.org/abs/2502.08524

310

Donatella Genovese @donatellag.bsky.social · 09/02/2025

A very interesting work that explores the possibility of having a unified interpretation across multiple models

021

Reposted by Donatella Genovese

Simone Scardapane @sscardapane.bsky.social · 10/01/2025

*MoE Graph Transformers for Interpretable Particle Collision Detection* by @alessiodevoto.bsky.social @sgiagu.bsky.social et al. We propose a MoE graph transformer for particle collision analysis, with many nice interpretability insights (e.g., expert specialization). arxiv.org/abs/2501.03432

0124