Generative AI news and analysis on Web Pulse: 13314 curated articles tracking the latest Generative AI developments, tools, and research, updated continuously from vetted sources.
Researchers have developed Aryabhata 2, a reasoning-focused language model for competitive STEM examinations, trained via reinforcement learning on PhysicsWallah's internal question banks. The model outperforms its base …
Researchers have developed PrismFlow, a new flow matching method that uses Koopman-inspired dynamical experts to generate high-quality time-series data. The approach addresses spectral distortion and poor mode coverage i…
Researchers introduced Spectral Guidance, a framework that controls diffusion models by exploiting the geometric structure of noise corruption to identify a small set of informative features. The method projects guidance…
Researchers have introduced Micro-Macro Retrieval (M2R), a new framework designed to reduce hallucination in large language models during long-form text generation. The system addresses the problem of factual errors by e…
A new study from arXiv (2605.29027) found that tonal variations in prompts cause systematic but model-dependent accuracy shifts in large language models (LLMs) on objective multiple-choice questions. Testing four cost-ef…
Researchers have developed Guidance Contrastive Policy Optimization (GCPO), a new algorithm that assigns per-token credit by contrasting model predictions under positive and negative prompts, addressing the limitation of…
Researchers have developed GenesisFunc, an automated pipeline that uses a multi-agent framework to generate high-quality, diverse training data for large language model function-calling. An 8B parameter model fine-tuned …
Researchers have identified a critical underfitting problem in diffusion bridge models near the target endpoint, caused by a mismatch in noise levels between network input and regression target. The team proposes the Noi…
Researchers have developed Orthogonal Concept Erasure (OCE), a new method for diffusion models that removes unwanted content while preserving overall image generation quality. The technique uses multiplicative parameter …
Researchers have developed TRACE, a trajectory-aware AI agent that uses large language model reasoning to optimize drug lead compounds by treating tool selection as a sequential decision-making process. The system outper…
A study of ClinicalTrials.gov records found a marked increase in AI-related clinical trials over time, with recent growth in references to machine learning, deep learning, chatbots, GPTs, and large language models. China…
Researchers have developed SalsaAgent, a multimodal embodied language model that generates expressive, full-body salsa dance motions in response to a human leader and synchronized to music. The system extends a large lan…
Researchers have developed GAP3D, a diffusion-based method that aligns vision-language model latents directly to patch-level image embeddings, enabling frozen generative models to use VLMs as prompt encoders without expe…
A developer built a nine-agent AI development team as a Claude Code plugin, orchestrating specialized agents to take a feature request from requirements analysis through production deployment. The plugin, called claude-d…
In 2026, AI agents have moved beyond chatbots to autonomously execute complex tasks, with major companies like DBS Bank, Visa, and Microsoft deploying them in production for credit card transactions and supply chain mana…
Software developers are entering a "disposable software era" where AI-generated code should be treated as replaceable infrastructure rather than permanent craftsmanship, according to a new industry argument. The shift mi…
UiPath reported quarterly revenue of $418.4 million, up 17% year-over-year, and net income of $22.5 million, reversing a $22.6 million loss from the prior year, driven by adoption of its agentic AI products moving from p…
Nearly 40% of active Data Engineer job postings now mention AI, with 17.4% explicitly requiring generative AI skills like LLMs, RAG, and AI agents, according to an analysis of 6,736 listings on InterviewStack.io as of Ma…
A developer identified a "deploy tax" that prevents non-developers from publishing AI-generated code, as existing deployment workflows require Git and terminal knowledge. To solve this, the developer built HTML Deployer,…
Researchers tested whether large language models can convert reader highlights from articles into durable spaced repetition flashcards, finding that while AI can identify a highlight's intent, it cannot yet determine whe…