Practical token optimization technique for developers building cost-sensitive LLM applications.
Researchers tested 1866 telegraph compression techniques on modern LLMs, finding telegraphese-style prompting achieves 25-34% savings compared to codebook approaches.
NVIDIA's Nemotron model family achieved gold-medal level performance on both International Olympiad in Informatics and International Mathematical Olympiad through specialized fine-tuning recipes.
PINNs outperform classical finite difference methods in 5D quantum harmonic oscillator problems, reducing memory requirements from 1TB+ to practical levels while maintaining 0.1% accuracy.
Researchers present a deterministic polynomial-time algorithm achieving optimal O(√d) regret for online inverse linear optimization, solving an open problem.
OpenAI published 722 math papers solving 90 top open problems using an internal AI model, including a breakthrough 'Quasi-Riemann Hypothesis' result.
Researchers developed an information-theoretic framework to quantitatively measure when specific features are generated during diffusion model denoising, revealing representation differences across pixel, SDVAE, VAVAE, and RAE models.
