January 2024 Summaries

In this paper review, we discussed the recent release of Phi-2, a small language model (SLM) developed by Hugging Face and AI21 Labs. We covered its architecture, training data, benchmarks, and deployment options. The key takeaways from this research are: 1. SLMs have fewer parameters than large language models (LLMs), making them more efficient in terms of memory usage and computational resources. 2. Phi-2 is trained on a diverse range of text data, including synthetic math and coding problems generated using GPT-3.5. 3. The model demonstrates competitive performance on various benchmarks, such as MMLU, HellaSwag, and TriviaQA, while being smaller in size compared to other open-source models like LLaMA. 4. Deployment options for Phi-2 include using tools like Ollama and LLM studio, which allow users to run the model locally on their hardware or even host it as a server. 5. There is ongoing research into extending the context length of SLMs through techniques like self-context extension, which could lead to more advanced applications in the future.

A well-developed data strategy is crucial for businesses to manage their data assets effectively and achieve their objectives. Key considerations include defining a vision for data, assessing current data assets, determining the right technology and tools, and building a data-driven culture. Creating a data moat involves identifying valuable first-party data sources, enriching existing data with external sources, and prioritizing data engineering efforts. To measure the ROI of their data strategy, companies should run data teams as profit centers, build an incremental roadmap, and avoid overspending on infrastructure. A successful data initiative requires a cultural shift within an organization supported by strong leadership and a clear vision.

The top AI conferences in 2024 are expected to be highly attended by industry leaders, researchers, and practitioners from around the world. Conferences like AI Engineer World's Fair, Arize:Observe, Cerebral Valley, World AI Conference, and NVIDIA GTC will focus on various aspects of artificial intelligence, including generative AI, machine learning, data science, and more. These events provide opportunities for networking, skill-building, and staying up-to-date with the latest advancements in AI. Many conferences are expected to take place in major cities like San Francisco, New York, Paris, and Las Vegas, offering a platform for attendees to engage with leading experts, professionals, researchers, and entrepreneurs. The conferences cover various topics, including AI ethics, data analytics, machine learning, deep learning, computer vision, natural language processing, and more. They also provide hands-on workshops, training sessions, and networking opportunities, making them valuable resources for anyone looking to learn about AI or advance their career in the field.

The workshop "RAG Time! Evaluate RAG with LLM Evals and Benchmarking" by Arize AI provided valuable insights into Retrieval Augmented Generation (RAG) and its applications. RAG enhances the output of robust language models by leveraging external knowledge bases, ensuring more accurate and relevant responses. The five key stages in building a RAG pipeline are loading data, indexing, storing, querying, and evaluating. A code-along exercise was provided to build a RAG pipeline using LlamaIndex and Phoenix Evals for large language model evaluation. The code-along exercise demonstrated how to install libraries, import them, launch the Phoenix application, download, load, and build an index, query the index, evaluate the results, compute NCDG and precision at 2, log evaluations to Phoenix, and perform response evaluation. The RAG pipeline was evaluated using Phoenix LLM evals, demonstrating its retrieval performance and QA correctness. The evaluation results showed that the system is not perfect but can generate correct responses ~91% of the time with a Hallucinations score of 0.05.