ML Research Scientist @Spotify. Personalization & GenAI. Previously: @eurecat_news, @Huawei, @UnivHelsinkiCS

Barcellona, Spagna
Just 2 days ago, we posted the preprint on arXiv. What better way to share the news than getting @_akhaliq to tweet about it?🚀🎉 Check it out!
Spotify presents Personalized Audiobook Recommendations at Spotify Through Graph Neural Networks In the ever-evolving digital audio landscape, Spotify, well-known for its music and talk content, has recently introduced audiobooks to its vast user base. While promising, this
5
37
17,374
Francesco Fabbri retweeted
Want to do cheaper, statistically valid LLM evals? Enter BetterEvals. It optimizes for real API costs (cached input / input / output) and uses confidence sequences, so you can stop sampling anytime without breaking validity. betterevals.org/
1
3
270
Francesco Fabbri retweeted
The teams really pulled out all the stops this year. We hope you like it! ❤️
3
8
75
21,601
Francesco Fabbri retweeted
Excited to share our paper “Semantic IDs for Joint Generative Search & Recommendation” @ RecSys'25 🧠 Jointly fine-tuning embeddings for both tasks → shared Semantic IDs that work for search and recs ⚖️ 📦 No more task-specific trade-offs!
2
3
8
498
Francesco Fabbri retweeted
Happy to share our #recsys25 paper: “Evaluating Podcast Recommendations with Profile-Aware LLM-as-a-Judge”. 🧠 90 days of listening → natural-language user profiles → LLM judges alignment 📊 Aligns with human eval. With amazing Spotify co-authors. 📄 arxiv.org/abs/2508.08777
2
5
13
1,038
Francesco Fabbri retweeted
What if we could use off-the-shelf Multimodal Large Language Model to enrich current video recommendation models? This is what we asked ourselves in our recent #recsys2025 paper arxiv.org/pdf/2508.09789 🧵
1
2
6
435
Francesco Fabbri retweeted
Semantic IDs for Joint Generative Search and Recommendation @_Guz_ et al. at Spotify introduce a bi-encoder model fine-tuned on both search and recommendation tasks to obtain item embeddings, followed by construction of unified Semantic ID space. 📝arxiv.org/abs/2508.10478
2
11
700
Francesco Fabbri retweeted
Evaluating Podcast Recommendations with Profile-Aware LLM-as-a-Judge Spotify introduces a profile-aware LLM framework for evaluating personalized podcast recommendations using natural-language user profiles distilled from listening history. 📝arxiv.org/abs/2508.08777
6
21
1,223
Francesco Fabbri retweeted
Can we effectively detect covert Information Operations (IOs) that attempt to manipulate socio-political debates on social media? This is the focus of our work, "IOHunter: Graph Foundation Model to Uncover Online Information Operations", just presented at the @RealAAAI
1
5
11
604
Francesco Fabbri retweeted
Internship opportunity at Spotify! PhD students, come join our Tech Research Team in London. When: June-August 2025 Last day to apply is *February 6th, 2025* at 5 PM GMT. Apply here lifeatspotify.com/jobs/2025-…
5
10
1,382
Francesco Fabbri retweeted
What a Christmas gift🎁In our latest @RealAAAI paper, we integrate our coordination detection model with the power of LLMs & GNNs to enhance the generalizability of information operation (IO) detection in data-scarce scenarios and unseen IOs. Kudos to @FractalZucky and our🌠team!
"IOHunter: Graph Foundation Model to Uncover Online Information Operations" goes to @RealAAAI '25! This is the result of an incredible collaboration with @LucaLuceri @Fra_Fabbri and @emilio__ferrara Read the entire thread for a brief explanation and the link to the preprint.
2
11
711
🚀🚀🚀
"IOHunter: Graph Foundation Model to Uncover Online Information Operations" goes to @RealAAAI '25! This is the result of an incredible collaboration with @LucaLuceri @Fra_Fabbri and @emilio__ferrara Read the entire thread for a brief explanation and the link to the preprint.
1
2
308
Francesco Fabbri retweeted
I am recruiting 1-2 PhD students for Fall 2025 at Princeton to work on Comp Social Science/Societal Impact of GenAI/GenAI for SocSci I wrote a bit about my research flavor & interests here: manoelhortaribeiro.github.io… Deadline: December 15th (cs.princeton.edu/grad#prospe…) Please boost!
8
252
684
77,032
Francesco Fabbri retweeted
📢 I’m recruiting PhD students for Fall'25 in @Berkeley_EECS or @UCJointCPH! Please apply if you have related interests in AI for social good / networks / public health / social science See my website for more on my research and how to apply: serinachang5.github.io/ Please RT!
20
236
798
92,054
Francesco Fabbri retweeted
It was a true honour to present our research at Spotify during #RecSys2024. Sharing our work in such a unique theatre setting was an unforgettable experience. Thank you to everyone involved!
7
36
1,798
Francesco Fabbri retweeted
Our paper on refusal in LLMs is finally up on arXiv.
14
41
363
79,704
Francesco Fabbri retweeted
🎉Excited to share: "Algorithmic Drift: A Simulation Framework to Study the Effects of Recommender Systems on User Preferences". Joint work with @EricaCoppolillo, @simone_mungari, @EttoreRitacco, @Fra_Fabbri, @FractalZucky, @FrancescoBonchi arxiv.org/abs/2409.16478
1
5
24
1,509
Francesco Fabbri retweeted
Uncovering Coordinated Cross-Platform Information Operations Threatening the Integrity of the 2024 U.S. Presidential Election Online Discussion w/ @emilio__ferrara @FractalZucky @Federico_Cinus View our findings here: papers.ssrn.com/sol3/papers.…
1
9
28
2,080
Francesco Fabbri retweeted
🚨Hiring alert! My team @SpotifyResearch is looking to hire a research scientist with a machine learning background! Come help us improve personalization experiences by integrating innovative generative AI technologies into our recommender systems. lifeatspotify.com/jobs/resea…
5
76
352
40,829
Francesco Fabbri retweeted
Video data is widely available but has been difficult to analyze automatically Particularly, TV data is a rich and important source of information for millions. We have a paper at #KDD2024 making such data accessible. Paper: gvrkiran.github.io/content/T…
2
31
148
16,064
Super excited to welcome @kdd_news in Barcelona 🚀 I’ll be at the conference this week, so feel free to ping me to talk about GNNs, GenAI for recs or anything Barcelona related 🎉
Waiting for #kdd2024 what’s better than Barça 🔴🔵 game? Thanks @Fra_Fabbri @SpotifyResearch for the invite!
1
28
1,974