Gustavo Penha @_Guz_

Research Scientist @Spotify · Working with IR, RecSys, NLP · PhD from @tudelft · ex @AmazonScience · https://t.co/SMu8BlyfIb linktr.ee/guzpenha Holanda (Países Baixos) Joined January 2009

Tweets

650
Followers

827
Following

565
Likes

10K

Kamil Ciosek @MLciosek

2 weeks ago

For anyone worried their LLM might be making stuff up, we made a budget‐friendly truth serum (semantic entropy + Bayesian). See for yourself: youtube.com/watch?v=x_8ORG… Paper: arxiv.org/pdf/2504.03579

0 7 3 685 2

Aixin Sun 孙爱欣 @AixinSG

4 weeks ago

I doubt to what extent improvements on these datasets would translate to improvements in today's real-world recommendation settings. Reference: arxiv.org/abs/2508.19399…

1 4 10 879 0

Download Image

Gustavo Penha @_Guz_

2 months ago

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 13 974 3

Sumit @_reachsumit

2 months ago

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

0 2 10 652 5

Sumit @_reachsumit

2 months ago

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

0 6 22 1K 10

Sumit @_reachsumit

2 months ago

Describe What You See with Multimodal Large Language Models to Enhance Video Recommendations @denadai2 et al. at Spotify use multimodal LLMs to generate natural-language descriptions of video content for better recommendations 📝arxiv.org/abs/2508.09789 👨🏽‍💻huggingface.co/datasets/marco…

0 4 8 602 3

Marco De Nadai @denadai2

2 months ago

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 6 391 1

Download Image

Sumit @_reachsumit

2 months ago

Adaptive Repetition for Mitigating Position Bias in LLM-Based Ranking Spotify introduces a dynamic early-stopping method that adaptively determines repetitions needed for each ranking instance, reducing LLM calls by 81% while preserving accuracy. 📝arxiv.org/abs/2507.17788

0 2 10 805 7

Sumit @_reachsumit

3 months ago

Aligned Query Expansion: Efficient Query Expansion for Information Retrieval through LLM Alignment @adam_x_yang et al. leverage LLM alignment techniques to fine-tune models for generating query expansions that directly optimize retrieval effectiveness. 📝arxiv.org/abs/2507.11042

0 1 6 521 5

Sumit @_reachsumit

5 months ago

Contextualizing Spotify's Audiobook List Recommendations with Descriptive Shelves Spotify introduces a pipeline that generates personalized audiobook recommendations with descriptive shelves to help users explore content based on their interests. 📝arxiv.org/abs/2504.13572

0 4 15 592 4

Gustavo Penha @_Guz_

6 months ago

I am attending #ECIR25 at Lucca 🇮🇹 if you are interested and want to discuss this position!

Gustavo Penha @_Guz_

6 months ago

I am attending #ECIR25 at Lucca 🇮🇹 if you are interested and want to discuss this position! https://t.co/OU0QvseMI2

0 2 10 2K 3

0 0 12 541 0

Download Image

Gustavo Penha @_Guz_

6 months ago

We have an open research scientist position in our lab at Spotify, Personalization ! The areas of expertise are: Information Retrieval, Recommendation System, Language Technologies, Foundational Models, Generative AI Technologies, and Machine Learning. lifeatspotify.com/jobs/research-…