Sowmya V.’s Post

We are organizing a Birds of a Feather session focused on "Benchmarking LLMs for Low Resource Languages." In this session, we aim to discuss the unique challenges faced by low resource languages. Our discussion will center on several key issues. Please join us; you are all cordially invited. Key issues: - Overcoming computational challenges - Evaluating LLMs using standard NLP tasks and datasets - Developing effective datasets for low resource languages - Benchmarking tools Majd Hawasly, Shammur Absar Chowdhury, Sabri Boughorbel, Nadir Durrani Basel Mousi, Yassine Elkheir Would like to invite: Ahmed Abdelali Preslav Nakov 🤟 Sudipta Kar Tanmoy Chakraborty Prof. (Dr.) Amitava Das Sunayana Sitaram

We are organizing a Birds of a Feather session focused on "Benchmarking LLMs for Low Resource Languages." In this session, we aim to discuss the unique challenges faced by low resource languages. Our discussion will center on several key issues. Please join us; you are all cordially invited. Key issues: - Overcoming computational challenges - Evaluating LLMs using standard NLP tasks and datasets - Developing effective datasets for low resource languages - Benchmarking tools Majd Hawasly, Shammur Absar Chowdhury, Sabri Boughorbel, Nadir Durrani Basel Mousi, Yassine Elkheir Would like to invite: Ahmed Abdelali Preslav Nakov 🤟 Sudipta Kar Tanmoy Chakraborty Prof. (Dr.) Amitava Das Sunayana Sitaram

The more research I see in the NLP community - the more I am reminded about the value of good quality data. See the recent paper "Textbooks Are All You Need II: phi-1.5 technical report" (arxiv.org/abs/2309.05463) for a recent example.

🚀 Learn how to design retrieval augmentation for generative pipelines with Haystack 🚀 In this talk, Tuana Çelik from deepset will share examples of retrieval augmentation in Haystack and how you might design your own pipelines for specific NLP use-cases. Let’s see how you can combine a vector search engine like Milvus with a RAG pipeline designed for your use case, to achieve the best out of powerful LLMs. Register with the link below 🔽 https://lnkd.in/enjq_5MB

📚 Join Tuana Çelik from deepset in this upcoming webinar on Aug 10. Learn how to implement retrieval augmentation in Haystack and design custom pipelines for specific NLP use-cases. Register and discover the power of combining Milvus with LLMs to take your projects to new heights! 👉 https://bit.ly/3Y9Qflv #Haystack #Milvus #OpenSource

PyData Global is absolutely one of my favorite conferences. In September, I spoke at PyData Amsterdam about monitoring NLP and LLM models in production. They finally published all the recordings! Come check it out. I briefly covered: 🧠 Embedding drift detection 🔢 Tracking interpretable text descriptors like text length and sentiment 🚦Using regular expressions to validate outputs 💬Explaining drift through model-based drift detection 📊Detecting changes in multi-modal data https://lnkd.in/e9AigeJ7

Part two of our educational research study series, “From neural nets to agent frameworks - techniques and tools for solving your NLP problem” is now up on our blog. In this study, we explore the Hugging Face transformers library to see how transformer model architectures perform on our text classification problem. As a reminder, this series is aimed at software developers and therefore will be technical in nature. https://lnkd.in/gTCUXi3h

📚 Join Tuana Çelik from deepset in this upcoming webinar on Aug 10. Learn how to implement retrieval augmentation in Haystack and design custom pipelines for specific NLP use-cases. Register and discover the power of combining Milvus with LLMs to take your projects to new heights! 👉 https://bit.ly/46Wbtr9 #Haystack #Milvus #OpenSource

I'm happy to share that our work "Maverick: Efficient and Accurate Coreference Resolution Defying Recent Trends" has been accepted at #ACL2024 main conference! 📰 In this paper we propose Maverick, a novel Coreference Resolution framework, that challenges the dominance of large autoregressive models by achieving new state-of-the-art performance on multiple benchmarks with significantly fewer parameters, lower memory usage, and faster inference. ✍ Work conducted with Edoardo Barba and Roberto Navigli at Sapienza NLP. 👀 Stay tuned for paper, data and models which are coming soon!

This book is amazing. If you want to understand how you can use transformers to do Natural Language Processing. This is exactly what I was looking for. .