The Evolving Landscape of LLM Evaluation
This post explores problems contributing to a benchmark crisis in LLM evaluation and potential solutions.
Latest
Command R+
This post discusses Command R and Command R+, the top open-weights model on Chatbot Arena at the time of its release and highlights their RAG and multilingual capabilities.
True Zero-shot MT
This post discusses recent results on extremely long-context benchmarks, explores true zero-shot machine translation (MT), and considers how to teach LLMs a new language like humans.
Thoughts on the 2024 AI Job Market
This post discusses macro trends I observed regarding the AI job market in 2024 and the reasons I joined my new company.
The Big Picture of AI Research
This post gives an overview of the Big Picture Workshop at EMNLP 2023.
NLP Research in the Era of LLMs
This post discusses compute as the main constraint for doing research in NLP and highlights five key research directions that do not require much compute.
EMNLP 2023 Primer
An overview of EMNLP 2023 papers covering QA, instruction tuning, task adaptation, NLG evaluation, and multilingual models and datasets.
NeurIPS 2023 Primer
A round-up of 20 exciting NeurIPS 2023 papers related to LLMs.
An Overview of Instruction Tuning Data
This post covers a range of widely used instruction tuning datasets, as well as important characteristics of instruction tuning data and best practices for using the datasets.
Modular Deep Learning
An overview of modular deep learning across four dimensions (computation function, routing function, aggregation function, and training setting).
The State of Multilingual AI
This post takes a closer look at the state of multilingual AI. How multilingual are current models in NLP, computer vision, and speech? What are the main recent contributions in this area? What challenges remain and how we can we address them?
ACL 2022 Highlights
This post discusses my highlights of ACL 2022, including language diversity and multimodality, prompting, the next big ideas and keynotes, my favorite papers, and the hybrid conference experience.
ML and NLP Research Highlights of 2021
This post summarizes progress across multiple impactful areas in ML and NLP in 2021.
Multi-domain Multilingual Question Answering
This post expands on the EMNLP 2021 tutorial on Multi-domain Multilingual Question Answering and highlights key insights and takeaways.
natural language processing
Challenges and Opportunities in NLP Benchmarking
Over the last years, models in NLP have become much more powerful, driven by advances in transfer learning. A consequence of this drastic increase in performance is that existing benchmarks have been left behind. Recent models "have outpaced the benchmarks to test for them" (AI Index Report 2021)
ACL 2021 Highlights
This post discusses my highlights of ACL 2021, including challenges in benchmarking, machine translation, model understanding, and multilingual NLP.
language models
Recent Advances in Language Model Fine-tuning
This article provides an overview of recent methods to fine-tune large pre-trained language models.
transfer learning
ML and NLP Research Highlights of 2020
This post summarizes progress in 10 exciting and impactful directions in ML and NLP in 2020.
cross-lingual
Why You Should Do NLP Beyond English
7000+ languages are spoken around the world but NLP research has mostly focused on English. This post outlines why you should work on languages other than English.
advice
10 Tips for Research and a PhD
This post outlines 10 things that I did during my PhD and found particularly helpful in the long run.
natural language processing
10 ML & NLP Research Highlights of 2019
This post gathers ten ML and NLP research directions that I found exciting and impactful in 2019.
cross-lingual
Unsupervised Cross-lingual Representation Learning
This post expands on the ACL 2019 tutorial on Unsupervised Cross-lingual Representation Learning. It highlights key insights and takeaways and provides updates based on recent work, particularly unsupervised deep multilingual models.
transfer learning
The State of Transfer Learning in NLP
This post expands on the NAACL 2019 tutorial on Transfer Learning in NLP. It highlights key insights and takeaways and provides updates based on recent work.
events
EurNLP
The first European NLP Summit (EurNLP) will take place in London on October 11, 2019. It is an opportunity to foster discussion and collaboration between researchers in and around Europe.