Full-text search across 9702 articles. Combine with topic and date filters; results sorted by relevance.
GGUF, GPTQ, and AWQ are the three dominant formats for running quantized large language models locally, each optimized for different hardware and use cases. GGUF, the format used by llama.cpp and its derivatives, offers …
Researchers introduced CoopSR, the first benchmark for multi-robot cooperative spatial reasoning, along with the EgoTeam dataset containing 114,227 question-answer pairs across 19 question types. The team also proposed S…
A group of 49 mathematicians compiled a dataset of 100 research-level mathematics questions with known answers during a workshop at the Max Planck Institute for Mathematics in the Sciences in Leipzig, Germany, between Ap…
Researchers have established a mathematical correspondence between hierarchical decision trees and diffusion processes, unifying the two model classes under a shared optimization principle called Global Trajectory Score …
More than 100 engineers cloned a free open-source handbook on large language model inference within its first week of release. The guide consolidates years of production experience and research to address the unique chal…
Two developers, Francis Tran and Elmar Chavez, built ClassifierAI, a prototype Chrome extension that automatically detects AI-generated content on the DEV platform. The extension analyzes an entire article and its cover …
A new study analyzing six large language models on the SWE-bench Verified benchmark found that agent-written tests do not significantly improve issue resolution rates, despite being a common practice. Researchers observe…
Researchers introduced Tangram, a serving system that enables non-uniform Key-Value cache compression for multi-turn large language model inference. The system uses deterministic budget allocation, head group page cluste…
Researchers have developed Discrete Tilt Matching (DTM), a likelihood-free method for fine-tuning masked diffusion large language models using reinforcement learning. The approach recasts fine-tuning as state-level match…
NVIDIA, startup Span, and homebuilder PulteGroup partnered in May to deploy XFRA nodes — liquid-cooled, fanless data center units containing 16 NVIDIA GPUs — mounted on residential exterior walls. The units tap unused el…
Researchers have developed a Bayesian statistical framework for migrating production systems reliant on large language models when the underlying model reaches end-of-life. The approach calibrates automated evaluation me…
Neo4j acquires GraphAware, integrating its Hume platform to offer governments an open-standards alternative to proprietary tools like Palantir Gotham. The company also launches Neo4j Virtual Graph for zero-ETL querying o…
A new AI safety research team within Arcadia Impact in London has formed over the past four months, growing to eight members who collaborate with the UK AISI alignment team. The team, led by research lead Andrew Draganov…
Seattle city council committees unanimously approved a one-year moratorium on new large-scale data centers, with a full council vote scheduled for June 9, 2026. The pause, paired with a resolution to draft regulations on…
A new empirical study of Pearl's Proof-of-Useful-Work blockchain found that its network of approximately 320,000 GPU-equivalents, consuming an estimated 112 megawatts of power, produced zero useful AI computation. Resear…
Amazon engineers publicly criticized the company's $200 billion data center spending spree, calling out the hypocrisy of massive AI infrastructure investment while simultaneously laying off 30,000 corporate employees. Sp…
Benedict Evans told Lenny Rachitsky that AI has reached its "1997 internet moment," representing a transformation as significant as the early internet but no larger. Rich Mironov warned that AI enables teams to ship 100x…
Researchers introduced generic triple-latent sequence models that maintain a running token state and compressed pair-memory pathway to capture higher-order token interactions without benchmark-specific parsing. The tripl…
Researchers have introduced On-Policy Representation Distillation (OPRD), a method that aligns student and teacher model representations across selected layers during training, bypassing the language model head to elimin…
Researchers have developed a post-training procedure called Latent Agents that distills multi-agent debate into a single large language model, achieving comparable or superior reasoning performance while using up to 93% …