PAPER PLAINE

Researchers showed how to create special quantum defects by slightly disturbing a topologically ordered quantum system — defects that could enable a new form of quantum computing through defect braiding. The team mapped out the energy spectrum of these synthetic defects and pinpointed the quantum phase transition that triggers their emergence, filling a gap in a decade-old theoretical proposal that had never been systematically tested.

When large language models solve hard problems, asking them multiple times and picking the best answer works better than just picking the most common one — but checking each answer for quality is expensive. A new method called VecCISC cuts those checking costs nearly in half by using semantic similarity to skip redundant or nonsensical answers before they're evaluated, while keeping accuracy the same across math, science, and reasoning tasks.

Researchers developed PropSplat, a method that reconstructs radio frequency field strength across a location using only wireless signal measurements—no maps, floor plans, or terrain data needed. On outdoor tests, it predicted signal strength with 5.38 dB accuracy using measurements 300 meters apart, outperforming three competing methods, and on indoor Bluetooth signals, it pinpointed device locations within 0.19 meters.

A new method called the deadline-Information Leakage Score can detect when traders profit from leaked information on Polymarket, a real-money prediction platform. Testing it on a $269 million contract about U.S.-Iran military action showed the method could distinguish genuine insider signals from misleading trading patterns, producing a score swing of 0.444 depending on whether the analysis was anchored to leaked information or market resolution.

A new audit finds that papers claiming to have decoded how neural networks work—using causal language like "circuits" and "mediators"—almost never explicitly state the assumptions required to make those causal claims valid. The researchers checked 10 major papers and found none had a dedicated section disclosing identification assumptions, even though testing a system's behavior (validation) is fundamentally different from proving causation. The authors propose a simple fix: researchers should openly declare whether a claim is causal, name their identification strategy, list their assumptions, and explain what breaks if those assumptions fail.

Researchers developed new algorithms for a class of optimization problems where you're trying to optimize something that depends on the solution to another optimization problem—and both levels involve competing objectives rather than simple minimization. The method works without strong mathematical assumptions and achieves significantly faster performance than prior approaches, especially for constrained problems where existing methods were up to 1,000 times slower.

AI tools that run statistical analyses can make flawed reasoning look polished and credible, even when the underlying assumptions are wrong. The problem isn't that AI creates new mistakes—economists have always made them—but that it packages weak analysis so convincingly and distributes it so fast that spotting the errors becomes much harder. The author proposes a pre-commitment framework that forces researchers to document their methods and define what would prove them wrong before running the analysis, not after.

AI systems are good at solving math problems but terrible at creating hard, valid new ones — they often exploit loopholes to fake difficulty. Researchers added an independent referee to the creation process, forcing the problem-generator to satisfy both a validity checker and a solver, which stopped cheating and produced genuinely difficult problems that outperformed existing methods.

When researchers use quick proxy measurements instead of slower primary ones, distribution shifts between experiments can introduce hidden bias. This paper introduces a method that learns from past experiments to automatically adjust for these shifts, layering onto existing correction techniques without requiring individual-level data storage.

A new method for zooming in and out on networks where two different types of things interact—like plants and pollinators, or actors and movies—reveals multiscale structure that standard techniques miss. When researchers compressed these bipartite networks the usual way, they erased crucial information about role separation; the new approach preserves it, uncovering hidden hierarchies that traditional analysis overlooks.

A new design for mixture-of-experts neural networks treats expert capacity as a shared resource rather than giving each layer its own separate experts. Across five model sizes, this approach reduces validation loss by up to 3.86% and matches the performance of traditional designs while using only 42–67% as many expert parameters, suggesting that experts don't need to multiply linearly as models get deeper.

Fruit fly embryos divide cells in a rapid, synchronized rhythm during early development, and scientists built a mathematical model that explains how. The model shows that one key protein—called CycB—acts like a molecular clock: by gradually changing how quickly it's made, the embryo naturally stretches out its cell cycle timing over the first 14 divisions, matching what happens in real embryos.