AI
Microsoft's Temporal GraphRAG Gives Changing Data a Clock
A newly published application organizes retrieval by time period and graph community, then routes a question to the relevant slice of a changing dataset.
Latest Articles
24 articles indexedAI
UiPath's Agentic-Loop Filing Lets Agents Build and Execute Tools
The pending application combines an API definition with an implementation, while related filings add human escalation, security evaluation and automation healing.
Tech Brief
AI
The module that sits between the camera and the language model
A newly published application describes a video-based multimodal system with a third component wedged between the image encoder and the language model. The placement is the whole invention.
AI
A Licence Key That Lives Inside the Generated Output
A Sony application published July 16, 2026 describes a generative AI system that embeds the authentication key into the content it produces — and makes the output measurably worse when that key is invalid. The licensing check stops being a gate and becomes a property of the artifact.
Artificial Intelligence
Replacing the Multiplier With a Table: Inside a Published Application on Lookup-Table Matrix Multiplication
A patent application published July 16, 2026 describes splitting an n-bit weight matrix into n one-bit matrices and precomputing every activation-group permutation into a lookup table, so quantized model inference becomes retrieval and addition rather than multiply-accumulate. The disclosed target is edge deployment on ordinary CPUs.
Artificial Intelligence
Telling an AI Agent What It's Allowed to Touch — in Plain English
A patent application published July 2, 2026 describes giving a machine-learning agent its permissions as a natural-language 'semantic entitlement,' then using a generative model to resolve which files and resources that actually covers. Here is how the mechanism works.
Artificial Intelligence
How NVIDIA's Invertible Neural Network Turns a Spectrogram Back Into Speech
A patent granted June 30, 2026 describes a flow-style vocoder that runs the same network two ways: training maps recorded speech to Gaussian noise, and inference maps noise back to a waveform. Here is how the mechanism works, and where it sits among a wave of generative-media grants issued to NVIDIA the same day.
Artificial Intelligence
A Neural Network That Trains Itself in the Circuit: Reading an Analog-Hardware Filing From This Week's Pub Drop
A newly published application describes a neural network built as an analog electrical circuit that learns by clamping its inputs and outputs, letting the circuit settle, and reading weight updates off the hardware itself. It is one of roughly 97 machine-learning applications in this week's drop, and a useful window into where analog AI hardware sits in the field.
Artificial Intelligence
What the NIST AI Risk Management Framework Is — and What Its Four Functions Ask You to Do
The NIST AI RMF is a voluntary framework for managing the risks of AI systems, organized around four functions: govern, map, measure, and manage. What the document says, in its own terms.
Artificial Intelligence
What Patent Class Are AI Inventions Filed Under? The Short Answer Is G06N
Most machine-learning and neural-network patents carry the Cooperative Patent Classification subclass G06N. What the code means, what its subgroups cover, and why it is the entry point for reading the AI patent landscape.
Artificial Intelligence
What 'Data Center Capex' Actually Means — and Why It Is the Number the AI Story Now Turns On
Capital expenditure on data centers is the spending hyperscalers report for building AI capacity. It is disclosed plainly in filings, and its trajectory is the defining figure of the AI buildout. What the term means, read through one company's 10-K.
Artificial Intelligence
What HBM Is — and Why High-Bandwidth Memory, Not Raw Compute, Often Bottlenecks AI
High-bandwidth memory stacks DRAM dies beside the processor to feed it data faster. It exists to relieve the memory-bandwidth bottleneck that limits AI accelerators. What HBM is, and why it shows up in the supply story.
Artificial Intelligence
Training vs. Inference Compute: The Two AI Costs That Behave Nothing Alike
Training builds a model; inference runs it. One is a large, one-time capital event; the other is a recurring per-query operating cost. Why the distinction decides how AI is priced, served, and disclosed.
Artificial Intelligence
What RLHF Is — How Human Feedback Turns a Raw Language Model Into One That Follows Instructions
Reinforcement learning from human feedback is the fine-tuning step that makes a large language model do what a user actually asks. Here is the mechanism, traced to the InstructGPT paper that demonstrated it.
Artificial Intelligence
What a Transformer Model Is — and Why 'Attention' Replaced the Recurrence It Was Built On
The transformer is the architecture under nearly every modern large language model. Its defining move was to drop recurrence and convolution entirely and rely on an attention mechanism. Here is what that means, from the 2017 paper that named it.
Artificial Intelligence
What a Mixture-of-Experts Model Actually Does — and Why It Lets a Model Grow Without Growing the Bill
A mixture-of-experts layer routes each input to a small subset of specialized sub-networks instead of running the whole model. Here is the mechanism, traced to the 2017 paper that made it work at scale.
Artificial Intelligence
ScholarQuest Benchmarks LLM Search Agents on Academic Paper Retrieval, Reporting a Best Recall@100 of 0.314
A new arXiv preprint introduces a taxonomy-guided benchmark for agentic academic paper search built from more than 1,000 computer-science topics and four research intents. The authors report agentic methods beat single-shot retrieval while leaving large gaps in recall.
Artificial Intelligence
AutoPass Puts an LLM Inside the Compiler to Tune LLVM, Reporting 1.04x and 1.12x Geometric-Mean Speedups Over -O3
A new arXiv preprint describes a multi-agent framework that lets a language model query LLVM's internal optimization state and use runtime measurements to refine compiler options. The authors report it operates without training or fine-tuning.
Tech Brief
Splitting a Tensor Without Breaking the Math: A New Tiling Method for Dataflow AI Chips
A published application describes how to carve a convolution's input tensor into memory-sized tiles and rebuild the result exactly, by computing the overlap each tile must carry from its neighbors. It sits inside a wider cluster of reconfigurable-dataflow filings published the same week.
Artificial Intelligence
Before It Writes a Rule, the GSA Is Asking the Obvious Question: What Happens to Federal Data Inside an LLM?
The General Services Administration is floating a draft acquisition clause on safeguarding data inside large language model systems — and, unusually, asking for comment before it commits to rulemaking.
Artificial Intelligence
To Catch a Multi-Stage Attack, Researchers Built the Dataset Nobody Had — and Fine-Tuned Small Models on It
A new arXiv release combines system, network, and browser logs with per-entry MITRE ATT&CK labels, then shows that small language models fine-tuned with LoRA can learn to read across all three.
Artificial Intelligence
Hype Check: 'Distilled' Datasets Were Supposed to Beat Just Picking Good Examples. A New Study Says They Don't.
Dataset distillation promises to compress huge training sets into a few synthetic images. A large arXiv benchmark finds that on real datasets, simply selecting a subset of real samples is competitive — and far cheaper.
Artificial Intelligence
The Real Bottleneck in Health AI Isn't the Model — It's Grading the Model
A new arXiv framework called RubricsTree replaces vague 'LLM-as-a-judge' scoring of health agents with 100-plus clinically verifiable yes/no checks, and reports large gains when those checks become training signal.
Artificial Intelligence
A Transformer That Keeps Thinking Until It Stops Changing Its Mind
FPRM, a new arXiv model, loops a Transformer over its own output and halts when the answer stops moving, letting it spend more compute on hard problems and less on easy ones.