Discrete diffusion, including remasking and uniform-state samplers, generate a sequence by writing multiple token positions per step, drawing each from a per-position distribution and choosing which positions to write from those same distributions. For domains of general interest (pixels, phonemes, ...
Extreme space-weather events can damage power systems on Earth and degrade GPS accuracy and satellite operations. A new machine learning system can predict where damage is likely to occur 30-60 minutes before a storm arrives.
The post Forecasting space weather risks on power grids appeared first on ...
arXiv:2609.35790v1 Announce Type: new
Abstract: While neural theorem provers have achieved impressive milestones in formal mathematics, they largely operate on the assumption that faithful Lean 4 formal statements are already provided. Translating informal natural language into a formal language is...
On the Effectiveness-Fluency Trade-Off in LLM Conditioning: A Systematic Study
Controlling the output of Large Language Models (LLMs) is a central challenge for their reliable deployment, yet a clear understanding of the involved trade-offs remains elusive. Current approaches to conditioning are often evaluated with a narrow focus on their effectiveness at injecting or removin...
When the Preconditioning Exponent Turns Negative: Learning-Rate Coupling and Cross-Environment Generalization
arXiv:2609.30271v1 Announce Type: new
Abstract: Adaptive optimizers are commonly parameterized by a fixed power of the second-moment estimate. Existing partially adaptive methods study exponents between momentum-like updates and the standard Adam square root, while the interaction between this expo...
Cosine Similarity Is Not Evidence: Measuring the Noise Floor of Interpretability Transfer Under Quantization
arXiv:2609.30275v1 Announce Type: new
Abstract: A statistic reported without the quantity needed to interpret it is not evidence. We develop that thesis for a concrete practice in AI safety. Interpretability artifacts are calibrated on full-precision weights, deployed on quantized ones, and certifi...
The Communication Bottleneck: A Round-Trip Study of Tree-Structured Expression Serialization in Language Models
When language models reason in chain-of-thought or exchange free-text intermediates, they serialize structured information into natural language. How much tree-structured compositional content survives this bottleneck? We propose a round-trip protocol that answers this question empirically for tree-...
One year in: How Microsoft Research Asia – Singapore is advancing research, partnership and talent for real-world impact
Since launching a year ago, the Microsoft Research Asia — Singapore lab has established a strong foundation, deepened collaboration across government, academia, and industry, and explored how frontier AI research can create real-world value.
The post One year in: How Microsoft Research Asia – Singap...
Quantum computing’s “dark horse” just proved it can go universal
Researchers have shown that exotic quantum particles called non-Abelian anyons can perform the full range of operations needed for universal quantum computing. Using 54 qubits on Quantinuum’s H2 processor, they combined braiding and fusion to unlock capabilities that braiding alone could not provide...
Why compounding containment is the real test of your AI strategy
I've been working in the AI space for about six and a half years. And one thing I can tell you with certainty is that the hardest part of any AI journey isn't getting started. It's what happens after the easy wins are gone.
The prompt that separates AI leaders from AI hobbyists
Three in four board members rate their AI knowledge highly, and their CEOs disagree. The gap shows up in a single prompt, and the fix takes about 30 seconds of extra typing...
CFD Correction of Open Tip Clearance Flow in a Compressor Cascade Using VAE Latent Space Adaptation
arXiv:2609.28558v1 Announce Type: new
Abstract: CFD predictions of open tip clearance flow in compressor cascades are subject to discrepancies relative to experiments, while experimental observations are sparse and high-resolution experimental ground truth is unavailable. This study proposes a non-...
arXiv:2609.28554v1 Announce Type: new
Abstract: We introduce the Pistis model family, comprising 27B- and 9B-parameter multimodal large language models built on Qwen3.6 and Qwen3.5, respectively, and developed through a general and scalable post-training framework. The framework first establishes a...
The Drift Contract: Spectral Updates for Depth-Robust Local Learning
arXiv:2609.26811v1 Announce Type: new
Abstract: Local learning trains each layer with its own auxiliary loss and no global backward pass, which makes layer updates structurally parallel. Two problems have kept it marginal: accuracy degrades as depth grows, and hyperparameters are fragile. We apply ...