
Giuseppe Colavito
@colavito_g
PhD Student @ UniBA 🇮🇹 | Collab Research Group
ID: 1002523494993793029
https://peppocola.github.io/ 01-06-2018 12:13:21
59 Tweet
138 Followers
272 Following

Our PhD student Giuseppe Colavito (Giuseppe Colavito) under the guidance of Prof. Nicole Novielli (Nicole Novielli) and Prof. Filippo Lanubile (Filippo Lanubile), will collaborate with Nasa to classify defect reports in the software created to prepare for future missions to Mars!🚀




Looking forward to your NLBSE-workshop tool competition entries on code comment classification, due **Dec 9** We give a larger dataset, a new scoring formula, and a base Colab notebook to train/test your solutions. See: nlbse2024.github.io/tools/ Questions? Contact us!

Exciting News for NLBSE-workshop'24 attendees! We are delighted to announce our keynote speakers; Albert Ziegler from GitHub and Michael Pradel from the University of Stuttgart. Stay tuned for a sneak peek into the topics they'll be covering! #KeynoteAnnouncement #ICSE2024


A little less than 2 weeks (due Saturday, Dec 9) to submit your NLBSE-workshop tool competition paper! Looking forward to your submissions for both issue and code comment classification tasks. More info: nlbse2024.github.io/tools/

Gentle reminder 🗣️ There is less than one week left to submit your work to the 3rd edition of the NLBSE-workshop workshop co-located with ICSE '24! We (Sebastiano Panichella @ Andrea) look forward to receiving your full, short, replication, & position papers by **Dec 7th, 2023**.

Our paper, "Leveraging GPT-like LLMs to Automate Issue Labeling," has been accepted at MSR 2025 !!! See you in Lisbon 🇵🇹!!! preprint: shorturl.at/bfpsA Nicole Novielli Filippo Lanubile Luigi Quaranta

SETFIT or GPT for issue labeling? Giuseppe Colavito at MSR 2025 argues that GPT is comparable to SETFIT and doesn’t have to be retrained. Joint with Nicole Novielli Filippo Lanubile and Luigi Quaranta


NLBSE-workshop End of the tool competition session. Congrats Kyle Luster Gabriel Aracena . Thanks Rafael Kallis , Giuseppe Colavito Ali Al-Kaswan, Luca Pascarella Oscar Chaparro, Pooja Rani





Excited to share that Giuseppe Colavito gave a talk at IBM on issue report classification! 🎤 He compared BERT-based models with GPT-like models, discussing trade-offs in performance, resource use, and deployment costs.





🚀 New paper alert! We benchmark several generative LLMs & BERT-like models on GitHub issues classification to uncover the trade-offs between accuracy, cost & practicality. 🔗 doi.org/10.1016/j.infs… #LLM #SoftwareEngineering #GitHub #NLP #AI Nicole Novielli Filippo Lanubile