Shahrukh Khan

Kolmogorov–Arnold Networks seem like a paradigm shift in ML architectures. Curious to see how they evolve. "While MLPs have fixed activation functions on nodes ("neurons"), KANs have learnable activation functions on edges ("weights"). KANs have no linear weights at all -- every weight parameter is replaced by a univariate function parametrized as a spline. " https://lnkd.in/gdPk_qCQ

29

Shashank Sonkar and colleagues are back with another great paper and matching GitHub repo, "Pedagogical Alignment of Large Language Models". You can read the paper on arXiv here - https://lnkd.in/enZfCJgi, and browse the code here - https://lnkd.in/gyW5hSzp. They demonstrate an interesting approach to generating synthetic data to train open weights models to behave more pedagogically, as well as developing some new evaluation metrics. I think there's lots here for those of us interested in this kind of work to learn from and build on. (If synthetic data generation is new do you, check out this post from answer.ai earlier this week - https://lnkd.in/ecsG7R4b.)

38

4 Kommentare

When *open* isn't *open source* 🙋🏼‍♂️👀 Reflecting on the open-source landscape for Large Language Models, I explore the essentials: permissive licensing, transparency in training data and model architecture & more #LLM #LLMs #AI #GenerativeAI #opensource https://lnkd.in/gSJuK_J4

1

[Call for Participation: 3 ICML paper presentations and 1 KDD Tutorial presentation] We are happy to share that our group is associated with the three ICML papers in the upcoming conference and is presenting a KDD Tutorial on Reasoning and Planning with Large Language Models in Code Development. ICML papers: [1] Collage: Light-Weight Low-Precision Strategy for LLM Training: We develop the training recipe of leveraging low-precision floating point units (FPUs) effectively provided the accounted error is properly compensated at the critical locations in the training process. https://lnkd.in/gVT8TS-x. [2] Variance-reduced Zeroth-Order Methods for Fine-Tuning Language Models: we leverage (forward pass only) zero-order methods with variance reduction techniques to enhance stability and convergence for LLM training, in need of smaller memory requirement than backprop-based methods. https://lnkd.in/gnwwanV8. [3] Automated Evaluation of Retrieval-Augmented Language Models with Task-Specific Exam Generation. The team proposed a new method to measure the task-specific accuracy of Retrieval-Augmented Large Language Models (RAG). https://lnkd.in/gV_paB8c KDD Tutorial: Multi-Hop Reasoning is an emergent capability of LLM. Such reasoning capability is critical for AI assisted code development. In this tutorial we explore the multifaceted impact of LLMs on code development, surveying techniques for generating high-quality codes, automatically documentation creation, code debugging, code decomposition, creating comprehensive test cases, among others.

112

3 Kommentare

🚀 **Just Published: Unlocking the Power of Large Language Models with Chunking Techniques!** 🚀 🔍 Ever wondered how large language models (LLMs) handle long documents while keeping context intact? The answer lies in **chunking**! In my latest blog, I explore the key chunking methods that boost LLM performance, covering everything from **simple sentence splits** to **adaptive AI-based chunking**. With each technique, I’ve included **code samples** to help you put these methods into action! 💡 Whether you're working in **Generative AI**, **NLP**, or **text summarization**, this guide will help you unlock new potential with LLMs. 📈 **Curious to know how it works? Check out the full blog!** https://lnkd.in/gniDQYwW --- #GenerativeAI #AI #MachineLearning #LLM #NLP #AIResearch #DataScience #Chunking #TextProcessing #AIEngineering --- For maximum engagement, tag a few companies or groups if possible, like: - **@OpenAI** - **@Google AI** - **@Hugging Face** - **@DeepMind**    Excited to share these insights and can’t wait to hear your thoughts! #llm #openai #genai #chunking #langchain

12

2 Kommentare

Here’s a blogpost that breaks down our work on Attribute Controlled Fine-tuning for Large Language Models . We introduced a new method that trains an auxiliary model to control a specific attribute (in this case toxicity). The approach regularizes the LLM's training by penalizing deviations from the desired (non-toxic) distribution, using the auxiliary model trained alongside the core LLM. This work, published at #EMNLP2024, was led by our intern Tao Meng (UCLA) together with our collaborators Ninareh Mehrabi, Palash Goyal, PhD, Anil Ramakrishna, PhD, Aram Galstyan, Richard Zemel, Kai-Wei Chang and Rahul Gupta. Blogpost: https://lnkd.in/eRzvKSKi Paper: https://lnkd.in/eHNwgWAt

34

This article (https://lnkd.in/gmXK46tT) introduces a novel approach to honeypot systems using Large Language Models (LLMs) based on paper (https://lnkd.in/gUZ5iz5v). By fine-tuning LLMs on attacker data, we can create highly interactive and realistic honeypots capable of capturing nuanced attacker behavior and providing valuable insights into cyber threats. Key benefits: ✔️ Realistic Interactions: LLMs generate human-like responses, making interactions with attackers more authentic. ✔️ Adaptive Learning: LLMs continuously learn from interactions, adapting to new attack techniques. ✔️ Deep Insights: LLMs analyze complex language patterns for a deeper understanding of attacker motivations and strategies. Future directions: ✔️ Expanded training data ✔️ Alternative fine-tuning techniques ✔️ Real-world deployment ✔️ Knowledge graph integration ✔️ Rigorous performance evaluation

32

1 Kommentar

🧠 Reflecting on: Why Graph Technology for AI Reasoning 🎙️ Speaking at DSC DACH was an incredible experience. It was a unique opportunity to dive deep into one of my favorite topics — how Graph Technology enhances AI, particularly through RAG — and engage in discussions with brilliant minds. We tackled key challenges like explainability, privacy, ethical responsibility, and knowledge management automation. 💡 Key takeaways: - Enhanced contextual understanding: Graphs create richer AI models through interconnected relationships. - Improved retrieval: Graph RAG allows AI systems to access diverse, private data sources, reducing biases and improving decision-making. - Privacy protection: By structuring and retrieving data securely, Graphs ensure sensitive information isn’t exposed. - Increased explainability: Graphs shift AI away from fully "black-box" models, offering clearer and interpretable insights. - Automated knowledge management: Graphs help integrate human domain-expertise with processes data, preserving institutional knowledge, enhancing decision-making, and driving innovation. 📖 For a deeper dive into these concepts, check out my article: Why Graph Technology for AI Reasoning (https://lnkd.in/dHHbq3NN) Looking forward to bringing more data stories to life! 🚀 #DSCDACH2024 #AI #Graphs #GenerativeAI #RAG #EthicalAI #Explainability #KnowledgeManagement

118

5 Kommentare

Our code for RAG evaluation is now publicly available at Amazon Science website: https://lnkd.in/g-HKUNZZ. This is the implementation of our evaluation mechanism performed by scoring the RAG on an automatically-generated synthetic exam composed of multiple choice questions based on the corpus of documents associated with the task. #OpenSource #OpenSourcing #AmazonScience #ICML2024 #Evaluation #LLMEvaluation

213

6 Kommentare

In this Technical Deep Dive, we share our approaches in using Large Language Models (#LLMs) to make numerical-type data(sets) like DHS, JMP, HDX more accessible with the example of WASH data. In this 15 min. video Olivier Mills and team cover: 1️⃣ Basic scripting to extract core data from datasets from API endpoints 2️⃣ How to use LLMs to generate LLM-friendly schema definitions 3️⃣ How to structure a AI-friendly database store that can scale to thousands of datasets and millions of data indicators 4️⃣ How to use vector querying to retrieve relevant indicators from the store based off of a natural language (english) query 5️⃣ How to build a prompt that generates valid SQL query using LLMs with the above retrieved data schema. The various components form a single AI system combining data ingestion and data exploration using a software layer, a data layer and multiple LLM layers. ⏭️ Next steps will be to expand the datasets, automate other data source ingestion and further test the system with more complex plain english queries. We plan to incorporate this as a new feature for WASH AI after testing. Feel free to reach out if you would like a copy of the Notebooks, sample prompts and data schema definitions. We are #opensource The initial work was done in collaboration with Data for Good YYC (see https://lnkd.in/g3AijS93)

3

1 Kommentar

Retrieval-Augmented Generation (RAG) enables answering functional, technical and domain-related questions about large codebases using LLMs, thus supporting developers in onboarding, feature development, or error analysis. In traditional RAG, code fragments relevant to the query are selected using similarity search. However, this method can miss critical code segments that are not directly related to the query but are crucial for correctly answering the question. As part of her master's thesis at itestra, Helena is researching an extension of RAG with enhanced code retrieval capabilities based on LLM agents. This aims to deliver more precise code snippets, significantly improving the quality of the answers. Helena uses Llama 3 70B and runs the entire setup on-premise on our own Nvidia RTX 6000 Ada, making it suitable for projects where the source code should not be processed in the cloud. #SoftwareEngineering #CodeAnalysis #ArtificialIntelligence #GenerativeAI #RetrievalAugmentedGeneration #LargeLanguageModel

27

4 Kommentare

Training LLMs to use inference-time feedback using large scale RL. Makes even the 8B Llama3.1 beat GPT-4 on CodeContests, and SOTA with the 70B. Author summary: LLMs for code should do much better if they can iterate on tests -- but they don't. Our new work (RLEF) addresses this with execution feedback at RL training time to use execution feedback at inference time. Notably, RLEF models are very sample efficient for inference. Competitive programming questions are often approached by sampling a large number of candidate programs; we can reach SOTA with just up to 3 samples. https://lnkd.in/g9_m_3Xm

18

3 Kommentare

🚀 Looking to streamline fine-tuning for models like Llama 3, Mistral, and Mixtral? Our latest blog shows how Anyscale simplifies LLMs fine-tuning with Ray: 🔑 Key takeaways: • Scalable fine-tuning with LLM-Forge • Model Deployment with RayLLM Read more: https://lnkd.in/gQwD4A2U

34

In Daniel Klitzke's latest article, he shows how you can use the Huggingface Transformers and Sentence Transformers libraries to boost you RAG pipelines using reranking models. #LLM #MachineLearning

23

New Post: More Agents Is All You need This paper introduces a method that significantly enhances Large Language Model (LLM) performance by scaling up the number of agents using a sampling- https://buff.ly/44Lmmvi

1

I am a few days late reading this, but this is a really good guide on data validation using the new OpenAI's o1 model via reasoning. Data quality evaluation is critical to AI model trust and transparency. In the past, data quality validation is done through pre-defined rules and pattern matching. The o1 model can understand the context of data and validate the data quality throught reasoning. Wow. Next step, I can't wait to try this on security data. I will report back on the experiment. Yichen Jin Bo LEI Kabir Mathur David B. Cross Wayne Duso Suman Natarajan Andrew Smyth https://lnkd.in/gEMbm6Nd

33

5 Kommentare