B, a 3-billion-parameter AI model, is challenging OpenAI, Google and DeepSeek on math and coding benchmarks while reigniting the debate over AI scaling, benchmark gaming and small-model reasoning.

Code and data reproduction package for the paper IAM-1: An Artificial-Free Geometric Pivot Method for Linear Programming Feasibility, following a full recheck and regeneration of the paper's ...

Retrieval-augmented generation enhances the performance of AI agents by expanding their recall. It can do this in three ...

NVIDIA diffusion language model Nemotron TwoTower achieves 2.42x LLM inference throughput without a full retraining run, ...

Multi-agent AI agent personality shapes outcomes in collaborative and negotiation workflows but not in structured coding, ...

Replace high-dimensional weights with a compact trainable latent vector — 200×–500× fewer trainable params ...