The Archive

HG-CRC enforces per-group risk guarantees in LLM selective prediction, preventing systematic error over-exposure to subgroups.

EgoPlay is event-triggered video editor for egocentric streams using fine-tuned V2V diffusion on Ego4D dataset.

At TechCrunch Disrupt 2026, the Smart Systems Stage will be where energy, infrastructure, and technology collide, covering everything from fusion breakthroughs to the grid strain AI is putting on the entire economy.

BettiSplit uses persistent Betti topology to guide privacy-aware split learning against feature inversion attacks.

LOCKS compresses KV cache for long-context LLM decoding via page-local spectral summaries, reducing cache reads.

EchoBridge aligns ECG and echocardiography text for cardiac findings using shared-private projection and prototype calibration.

LLM-assisted workflow constructs French legal knowledge graph from maintenance regulations via ontology engineering.

This $9 NFC key requires you to physically scan it to unlock distracting apps on your phone.

Unsupervised sentence embedding methods (TSDAE, CSE) adapted for ancient language text analysis without labeled data.

Framework for auditing multimodal LLM faithfulness in power grid diagnosis via modality ablation and behavioral reliance testing.

K-SCAN hybrid clustering algorithm balancing DBSCAN robustness with K-Means scalability via vector quantization.

Best practices guide for ML-driven Earth observation geospatial map production addressing error propagation in pipelines.

FlowCTS extends on-policy distillation to flow models via trajectory supervision matching for improved generation.

After two years in stealth, Safe Superintelligence has announced a long-term partnership with Nvidia as it prepares to scale to its next phase.

Benchmark of six EEG foundation models (LaBraM, EEGMamba, CBraMod, REVE, BENDR, BIOT) on clinical tasks with negative controls.

GPU-accelerated symmetric NMF for large-scale dependence decomposition in portfolio risk estimation.

DecoupleMix formulates VLM pretraining data curation as mixture optimization, decoupling capability ratios from domain allocation.

Evaluation of multilingual NMT models (NLLB, mBART) on English-Tamil translation across four datasets with BLEU/chrF metrics.

Physics Transformer adapts attention for PDE solving via functional tokenization respecting infinite-dimensional field structure.

Robotics learns from bitter lesson scaling; AI agents complete week-long programming tasks; OpenAI discovers accidental vulnerability in reasoning model.

The massive seed round was led by Index Ventures and Ribbit Capital, with participation from Sarah Guo's Conviction Partners.

Nvidia on Monday said it is joining forces with Microsoft, SpaceX, IBM, and other tech companies to build and share open-source AI security tools. The new Open Secure AI Alliance said open tools are required to effectively defend against attacks from frontier models. The initiative is a direct response to mounting concerns over the safety of advanced AI systems after a rogue OpenAI model escaped containment and attacked another company during testing. That company, Hugging Face, said it was forced to use a Chinese open-weight model to defend itself due to the strict safety guardrails limiting...

Imagine a healthcare system made up of multiple AI agents: one that manages symptom assessment, another scheduling, a third insurance, and a fourth pharmacy. Each is an expert in its domain. But they all have their own distinct knowledge and objectives. Today they can exchange data, but they are not yet able to actually coordinate…

Drug discovery is a high-cost, high-risk endeavor that is under growing pressure from a market increasingly defined by first-mover advantage. Since the 1950s, the cost of developing new pharmaceuticals has roughly doubled every nine years—a phenomenon known as Eroom’s Law. Today, bringing a new drug to market takes an average of 10-15 years and costs…

For the enterprise, the promise of agentic AI is much more than just a better chatbot. It is software agents that execute business tasks end-to-end across people, business workflows, data, and systems. The platform best-suited to run agents is built with proper CPU capacity, resilient data access, policy-aware tool use, observability, memory management, and the…

Meme generator may have screwed up by using templates in outputs, expert says.

Stratechery taking editorial vacation week of July 27; publishing schedule suspended except Sharp China and Asianometry.

Building a great AI agent isn’t just about choosing the right models. The harness is the architecture surrounding the model. How it renders context, executes... Building a great AI agent isn’t just about choosing the right models. The harness is the architecture surrounding the model. How it renders context, executes actions, manages state, and decides when a task is done shapes outcomes just as much as the model itself. Harness design alone can account for double-digit swings in benchmark results and significant differences in token cost… Source

OpenAI research documents how ChatGPT users expand task scope across roles, with implications for job design and workforce skill requirements.