Nabeel Seedat (@nabeelseedat01) 's Twitter Profile
Nabeel Seedat

@nabeelseedat01

PhD student in Machine Learning @Cambridge_Uni | #DataCentricAI | 🇿🇦

ID: 1017347753536638977

calendar_today12-07-2018 09:59:40

96 Tweet

321 Followers

1,1K Following

Mihaela van der Schaar (@mihaelavds) 's Twitter Profile Photo

1) Today at #ICML2024! Join us in Session 3 (11:30 CEST) to talk about how “Curated LLM: Synergy of LLMs and Data Curation for tabular augmentation in low-data regimes" puts forward technical proposals to leverage LLMs for data augmentation. 1/2

1) Today at #ICML2024! Join us in Session 3 (11:30 CEST) to talk about how “Curated LLM: Synergy of LLMs and Data Curation for tabular augmentation in low-data regimes" puts forward technical proposals to leverage LLMs for data augmentation. 1/2
Mihaela van der Schaar (@mihaelavds) 's Twitter Profile Photo

Are you at #Miccai2024? Join us this Sunday for our tutorial 'Clinical AI in the Real-World: From Data-Centric AI to Dynamic Learning'! I'm excited to present with Nabeel Seedat & Camila González as we explore methods to make healthcare AI ready for the real world 🏥MICCAI Society

Open Life Science AI (@openlifesciai) 's Twitter Profile Photo

3/25 Matchmaker: Self-Improving Large Language Model Programs for Schema Matching This paper introduces Matchmaker, a compositional language model program for schema matching that addresses the challenges of structural and semantic heterogeneity in data sources. It uses a

3/25 Matchmaker: Self-Improving Large Language Model Programs for Schema Matching

This paper introduces Matchmaker, a compositional language model program for schema matching that addresses the challenges of structural and semantic heterogeneity in data sources.  

It uses a
Mihaela van der Schaar (@mihaelavds) 's Twitter Profile Photo

Join our #InspirationExchange session today about AI Model Evaluation! Nabeel Seedat & Paulius Rauba will present our upcoming #NeurIPS2024 work - learn how Context-Aware Testing ensures reliable ML: introducing a new paradigm for testing ML models. vanderschaar-lab.com/engagement-ses…

Join our #InspirationExchange session today about AI Model Evaluation! <a href="/NabeelSeedat01/">Nabeel Seedat</a> &amp; <a href="/PauliusRauba/">Paulius Rauba</a> will present our upcoming #NeurIPS2024 work - learn how Context-Aware Testing ensures reliable ML: introducing a new paradigm for testing ML models. vanderschaar-lab.com/engagement-ses…
Mihaela van der Schaar (@mihaelavds) 's Twitter Profile Photo

Introducing #CliMBDC! This LLM-powered copilot combines data-centric AI tools & advanced reasoning to solve complex data challenges in healthcare research & beyond. CliMB-DC can perform data management & build predictive models using natural language in a single conversation 1/2

Introducing #CliMBDC! This LLM-powered copilot combines data-centric AI tools &amp; advanced reasoning to solve complex data challenges in healthcare research &amp; beyond.

CliMB-DC can perform data management &amp; build predictive models using natural language in a single conversation 1/2
Mihaela van der Schaar (@mihaelavds) 's Twitter Profile Photo

Interested? Explore the architecture and capabilities of CliMB-DC on our website & watch our comprehensive engagement sessions where we demonstrate and discuss our copilot here: vanderschaar-lab.com/climb-dc/ You can access the CliMB-DC paper directly here: arxiv.org/pdf/2501.10321 2/2

Jiashuo Liu (@liujiashuo77) 's Twitter Profile Photo

Probably the 1ST data-centric LLM copilot, integrating various data-centric tools, supporting complex data processing and modeling of clinical problems🔥🔥 - Paper: Summarizes various data-related issues & corresponding tools, along with detailed case studies in the medical field

Jonathan Richard Schwarz (@schwarzjn_) 's Twitter Profile Photo

Scales++: New work at the intersection of efficient LLM Evals & Data Pruning using embeddings measuring the cognitive complexity of your eval data 🧠 📈 Get accurate evals using only 0.5% of your data with a reduction in upfront cost of over 18x! 👉 arxiv.org/abs/2510.26384

Scales++: New work at the intersection of efficient LLM Evals &amp; Data Pruning using embeddings measuring the cognitive complexity of your eval data 🧠

📈 Get accurate evals using only 0.5% of your data with a reduction in upfront cost of over 18x!
 
👉 arxiv.org/abs/2510.26384
Nabeel Seedat (@nabeelseedat01) 's Twitter Profile Photo

Do you need to use your full benchmark to evaluate your LLM? Check out our new work from Thomson Reuters Foundational Research on more efficient LLM evals