GenAI

It’s been a while since I last wrote a blog post, and I’m excited to be back!

We’re on a journey to advance and democratize artificial intelligence through open source and open science.

In this blog post we will build a local Deep research agent using Qwen 3 8b model, LangGraph, Composio, and Ollama.

An example showing how A2A and MCP can be used together - ishanExtreme/a2a_mcp-example

Visual RAG expands AI's ability to understand and utilize charts, graphs, and images, a critical skill as 65% of people are visual learners. Mastering this technology allows you to build truly multimodal AI systems that can reason about visual data, giving you a competitive edge in enterprise AI development and opening new possibilities for data-driven applications.

Explore the top Retrieval-Augmented Generation (RAG) techniques of 2025, including Traditional RAG, Long RAG, Self-RAG, and more.

The Contradiction Psb Lds model is a sentence-transformers model that maps sentences and paragraphs to a 768-dimensional dense vector space, allowing for tasks like clustering or semantic search. But what does that mean for you? It means you can use this model to identify contradiction sentences in patents with ease. It's built on top of the PatentSBERTa model and can be used with either sentence-transformers or HuggingFace Transformers. The model is efficient and can be used for a variety of tasks, but have you ever wondered how it was trained? It was trained with a batch size of 16 and a learning rate of 2e-05, with a total of 1128 steps per epoch. The model's architecture is based on the MPNetModel, which is a type of transformer model. So, what makes this model unique? It's ability to identify contradiction sentences in patents, making it a valuable tool for anyone working with patent data.

State of the Art Natural Language Processing. Contribute to JohnSnowLabs/spark-nlp development by creating an account on GitHub.

TL;DR Deep research is the hottest AI feature right now. Every LLM provider is adding to...

Streamline the process of working with legal contracts by implementing an end-to-end solution using agentic GraphRAG.

Vector search represents a revolution in information retrieval. Vector embedding - by taking account of context and semantic meaning - empowers vector search to return more relevant and accurate results. In this article we compare two common approaches to multi-attribute vector search.

Learn about Google's Agent2Agent (A2A), an open protocol for AI agent collaboration across different systems.

Weights & Biases, developer tools for machine learning

DeerFlow is a community-driven framework for deep research, combining language models with tools like web search, crawling, and Python execution, while contributing back to the open-source communit...

Natural language inference (NLI), also known as Recognizing Textual Entailment (RTE), is an important aspect of natural language understanding. Most research now uses machine learning and deep learning to perform this task on specific datasets, meaning their solution is not explainable nor explicit. To address the need for an explainable approach to RTE, we propose a novel pipeline that is based on translating text into an Abstract Meaning Representation (AMR) graph. For this we use a pre-trained AMR parser. We then translate the AMR graph into propositional logic and use a SAT solver for automated reasoning. In text, often commonsense suggests that an entailment (or contradiction) relationship holds between a premise and a claim, but because different wordings are used, this is not identified from their logical representations. To address this, we introduce relaxation methods to allow replacement or forgetting of some propositions. Our experimental results show this pipeline performs well on four RTE datasets.

A curated list of 100+ resources for building and deploying generative AI specifically focusing on helping you become a Generative AI Data Scientist with LLMs - business-science/awesome-generative-...

Explaining the Model Context Protocol and everything that might go wrong.

Learn common agentic design patterns and workflows for building robust, scalable AI applications, understanding when to use each.

Here we demonstrate how to pass multimodal input directly to models.

Learn about how Azure API Center can be a centralized registry for MCP servers in your organization. Developers and other stakeholders can use the API Center portal to discover MCP servers.

Explore this post and more from the LangChain community

Learn when and how to use GraphRAG and how it can improve on some search tasks

I recently published a blog post on how to get started with the official A2A demo. In it we explored the capabilities of A2A and how it helps AI agents, potentially built with different frameworks…

As AI Agents and assistants become increasingly central to modern applications and experiences, the need for seamless, secure integration with external tools...

This repository shares end-to-end notebooks on how to use various Weaviate features and integrations! - weaviate/recipes