Natural Language Processing news and analysis on Web Pulse: 2523 curated articles tracking the latest Natural Language Processing developments, tools, and research, updated continuously from vetted sources.
A developer provides a decision tree and code examples to distinguish between RAG (Retrieval-Augmented Generation) and agentic AI architectures. RAG is recommended for answering questions from documents, while agents are…
A developer built dspyer, an open-source tool that wraps LLM calls in Pydantic schemas to automatically correct malformed JSON, missing fields, or fabricated citations. The tool compiles to a standard DSPy module, enabli…
Serafis, a San Francisco-based AI startup founded in 2025, provides narrative intelligence by indexing and structuring conversations from podcasts and independent media into searchable data for investors and AI agents. B…
Gradium launched two real-time speech translation models, stt-translate and s2s-translate, covering 20 language pairs across English, French, German, Spanish, and Portuguese. The models collapse the standard three-model …
Meta announced a new line of AI glasses starting at $299, including a Kylie Jenner edition for $399, featuring the Muse Spark model, live translation in 20 languages, and a 3K camera. The move marks Meta's shift to own-b…
Apple's iOS 27 introduces several new AirPods features, including a custom EQ for H2-chip models, Siri AI integration via iPhone, Precision Finding for AirPods Pro 3 with Apple Watch support, GymKit syncing, and Name Rec…
A researcher has developed a technique called 'ontological inversion' that allows large language models to capture the multifaceted nature of complex concepts like sorrowfuljoy, rather than overfitting to singular emotio…
A developer built a production-grade API for a 3B text-to-SQL model using FastAPI and Ollama, enabling natural language queries against SQLite databases at zero inference cost. The API, part of the de-swarm project, incl…
A new service called ccMarvin lets users email marvin@ccmarvin.com to receive AI-powered research, portfolio tracking, and custom newsletters. The tool functions as a virtual research analyst, capable of monitoring news,…
Researchers from the UK and Germany identified three key drivers behind 'AI psychosis' in a study published in Nature's Digital Psychiatry and Neuroscience. The drivers—linguistic alignment, hyperpersonalization, and the…
Chunking strategies determine which slices of a document reach an AI model's context window, directly shaping retrieval accuracy. A 2026 benchmark found recursive 512-token chunking reached 69% accuracy versus 54% for se…
Google's Gemini desktop app adds a voice dictation feature called 'Speak to Window,' allowing users to hold the Fn key and issue voice commands across any open application. The feature, rolling out alongside 'Magic Point…
Mistral released OCR 4 on June 23, a document-processing model that returns structured data with bounding boxes, block types, and confidence scores, targeting enterprise back-office tasks like invoice processing and comp…
The Washington Post tested major AI chatbots for political bias, finding OpenAI's GPT-5.5 gave left-leaning arguments 80% of the time, while Google's Gemini presented balanced views in over 90% of responses. The evaluati…
Neonmem 0.9.7 introduces a two-level importer that separates folders and files into a searchable knowledge pool and agent chats into typed memories, using IBM Granite-30M embeddings via ONNX Runtime for offline, grounded…
Sentrup launched an AI-powered customer support platform that enables businesses to embed a chat widget into websites or mobile apps, providing 24/7 support by uploading a knowledge base document or URL, with a setup tim…
Hootsuite launched Social OS and Wisdom, a social-first AI agent, on June 24, 2026, reorganizing its product portfolio into four AI-powered tools connected by a shared agentic layer. The company also introduced MCP conne…
Kindway built Reach, an AI agent platform for SMBs on WhatsApp and Instagram, but found that non-technical users struggled with prompt engineering. The team pivoted from a prompt-editing interface to a feedback loop wher…
A new guide on evaluating Retrieval-Augmented Generation (RAG) systems identifies five common mistakes in current evaluation practices, drawing on insights from five research papers. The article advises practitioners on …
A developer built a local AI research assistant to brief them on investing, after concluding that LLMs are useful for research synthesis, comparative analysis, and thesis stress-testing but not for price prediction. The …