LLMs & Generative AI - Page 15 of 55
Latest breakthroughs in large language models and generative AI shaping the future of artificial intelligence and machine learning.
1086 articles
View complete article list
Proposal Calls for Data Probes to Study Impact of Training Data on LLMs
The way we train AI is still medieval. We throw it a dump truck of text and hope it learns something. We shovel the pile, sift it, then shrug when the model spits out nonsense.
Isotonic calibration gets O(n⁻¹/³) sample complexity, cost‑optimal LLM routing
Everyone wants to pay less for AI. The standard method is cascading: you send an easy query to a cheap, small model and only bother the expensive one with the hard stuff. The trick is guessing which queries are which.
Basis Spline Decoupling Enables Compression of Transformer Models
The best way to make a big AI model smaller is to pretend it's a different, simpler kind of mathematical beast. This trick is called decoupling, and it's been around.
LLM Retrieves Median 2020 Inflation Expectation, Drowning Prompt Guidance
The model knows the numbers. It has seen them before, during training, embedded in the weighty folds of its neural architecture. Ask for the median 2020 inflation expectation, and it doesn’t reason, it retrieves.
QuickReduce FP4 delivers ~4.1× speedup over RCCL at TP=4 for large messages
Training today's massive AI models creates a monumental traffic jam. The gridlock isn't the raw horsepower of the GPUs themselves. It's the agonizingly slow process of shuttling data between them, an operation called all-reduce.
Study Uses SHARP and New Error Framework to Assess PHRs in Health AI
Ask a health AI about a symptom. Should it consider the medication you started last Tuesday, or ignore it? Personal health records exist to provide that crucial context.
Google's Gemini 3.5 Flash, pricier, adds 11 Omniscience points hallucinations 61%
Google's Gemini 3.5 Flash just slashed its hallucination rate by a staggering 31 points. Don't break out the champagne yet, though. At 61 percent, it’s still hallucinating more than twice as often as front-runners like MiMo-V2.5-Pro.
Alibaba launches Qwen3.5‑LiveTranslate‑Flash: 60‑language translation in 2.8 s
The Qwen3.5‑LiveTranslate‑Flash doesn’t just translate sixty languages in under three seconds. It steals your voice. That’s the real trick.
I/O 2026 unveils Gemini Omni for universal creation, Gemini 3.5 Flash debut
A video is no longer just a video. It’s raw material, for anything you can imagine. At I/O 2026, Google unveiled Gemini Omni, a model that starts with video but doesn’t stop there.
EKS Hosts Multistage Multimodal Recommender; DLRM Personalizes Rankings
Personalization doesn’t require knowing who a user is. Even for the unknown visitor, served the same popularity-biased slate by the retriever, a DLRM ranker rewrites the rules.
Gemini 3.5 Flash Enhances Web UI, Graphics and AI Studio Animations
Google's latest demo isn't just another chatbot. It's a construction crew for your brainwaves. The company showed Gemini 3.5 Flash building interactive hardware simulators and full brand kits from a simple text prompt.
Fresh Web Data Grounds LLMs, Highlighting RAG's Production Limits
The promise of Retrieval-Augmented Generation was always compelling: ground your AI in trusted data, eliminate hallucinations. In theory, it works. In practice, a quiet crisis emerges. RAG systems are only as intelligent as their last update.
ANNEAL lets neuro‑symbolic agents patch knowledge graphs without weight changes
Our big AI models are stuck on repeat. They get something wrong, get corrected, and then get the exact same thing wrong all over again. It’s expensive amnesia. ANNEAL tries to stop that.
Claude Cowork: Guide to Turning Q1 Sales Data into a Structured Word Report
Q1 sales data is raw potential, a chaotic pile of numbers, trends, and insights waiting to be shaped into a narrative your stakeholders can actually use. The problem isn’t the data; it’s the process.
Activation steering reveals latent bias in LLMs, reinjection restores decisions
A language model can pass every fairness check and still be rigged. The real bias isn't in what it says, but in how it thinks, a hidden tilt in its internal wiring that only shows up when you push on the right spot.
95% of task‑specific generative AI pilots never reach production
We are very bad at turning AI prototypes into actual things people can use. The industry's own number says ninety-five percent of these small, task-specific generative AI projects die before they ever go live.
5 Practical Uses of Local Language Models Highlight Code‑First Approach
Local language models don’t need a cloud connection to do real work. I built a minimal Python agent running Llama 3.2 through Ollama, no external API, no fallback, and watched it plan, search, write files, and finish tasks on its own.
SkillSmith extracts fine-grained boundaries so agents run only needed components
Most AI agents today are overpacked. They lug their entire skill library into every minor task, dragging useless context along and burning expensive tokens on unnecessary thoughts. Enter SkillSmith, a method detailed in a recent paper.
Quantized LLMs Show Emerging Bias, Masking Gradual Degradation
Quantizing a model is like tightening a tourniquet. You're told it's just compression, that the core intelligence survives intact. The new numbers say otherwise.
AgentStop cuts GPU power, heat and battery drain by ending AI agents early
Your phone's battery is dying because your AI agent is doing too much thinking. Specifically, it's doing useless thinking, following doomed paths of logic until they crash into a wall. This process eats power and cooks the processor.
Featured Resources & Reviews
No Code MBA Course Review
Learn to build AI-powered apps without coding. Our comprehensive review of No Code MBA's course.
AI Tools & Resources
Curated collection of AI tools, courses, and frameworks to accelerate your AI journey.
Weekly AI Digest
Get the week's most important AI news delivered to your inbox every week.