Engineering
SearchAtlas: Analyzing Agentic Search Strategies via Evidential Query Graphs
SearchAtlas is a framework that converts raw search trajectories produced by LLM agents into structured evidence graphs showing how evidence…
Engineering
RAG-Safety-Bench: Reliable Evaluation of Retrieval-Augmented LLM Safety
The paper introduces RAG-Safety-Bench, a benchmark isolating how retrieval-augmented generation affects LLM safety by testing four controlled conditions: no-RAG baseline…
Engineering
LOCUS: Task-Aware Low-Rank Post-Training for Token-Efficient Language Generation
The paper introduces LOCUS, a post-training method that curbs verbosity in large language models by applying task-aware low-rank adaptations that…
Engineering
Structured Transforms for Low-Overhead Quantization of Language Models
Researchers present an enhanced quantization method for large language models that improves on Kashin-decomposition-based approaches by replacing dense random orthogonal…
Engineering
vLLM x AgentX: Optimizing for Real-World Agentic Serving
The post describes a three-layer stack vLLM uses to optimize serving for agentic workloads: a data plane with hybrid KV…
Engineering
Tiered KV Cache Offloading in vLLM
vLLM’s tiered KV cache offloading system preserves evicted key-value cache data across host memory, storage, and remote peers instead of…
Engineering
An Open Recipe for IMO Gold: Training Nemotron for Olympiad Mathematics
The paper details training Nemotron 3 Ultra with supervised fine-tuning and reinforcement learning to produce natural-language mathematical proofs, built around…
Engineering
A Fragility Spectrum for Recursive Language-Model Training
The paper studies how 13 different language model checkpoints degrade under recursive training on model-generated text, using a shared corpus…
Opinions
Understanding ‘AI psychosis’
This American Psychological Association Monitor on Psychology article, published in the September 2026 issue, examines the debate among researchers and…
Opinions
AI has finally come for math—mathematicians might never recover
Published around September 11, 2026, this Scientific American article examines how AI systems now capable of solving longstanding mathematical problems…
News
Anthropic Signs $35 Billion Cloud Deal With Nvidia-Backed Lambda
Anthropic has entered a $35 billion, 15-year cloud-computing agreement with Lambda, an Nvidia-backed provider, to secure capacity at a 350-megawatt…
Engineering
Helion x Hugging Face Kernels: Building and Shipping Out-of-the-box Performant Kernels
This PyTorch blog post details how the Hugging Face Kernels project added support for Helion, a tile-based DSL from Meta…