Preprint method matches exact quantum test and anthracene spectrum
A Hagedorn-wavepacket calculation agreed closely with a controlled harmonic-model result and an experimental spectrum, while revealing important limits.
1741–1760
A Hagedorn-wavepacket calculation agreed closely with a controlled harmonic-model result and an experimental spectrum, while revealing important limits.
A theorem by Michael Eichmair and Thomas Koerber gives a complete classification under Serrin’s stated equations, boundary conditions and stability requirement.
A theoretical framework replaces a difficult matrix calculation with scalar bounds, but its promise remains untested outside modelled quantum systems.
Simulator and field logs showed strong radius and driver signals, while curve direction was weaker and uphill behavior harder to model.
A deterministic formula links the difference between chain and order polytopes to star elements and extends to a specified marked chain–order construction.
A nine-language benchmark found the steepest gaps in Swahili and Zulu, while proprietary systems were more stable than open-source and medically specialized models.
The continual-learning defense had lower benchmark attack success rates than SecAlign and LlamaGuard while matching or exceeding the undefended base model on two question-answering tests.
A 3-D simulation of the Lesser Antilles showed a temperature rise above 226 K, a ridge emerging after about 5 million years and modeled mantle melting 20–50 km down.
A resonant-state expansion matched Mie theory and COMSOL in modeled cases while revealing frequency-selective routing in an annular metaatom.
A theorem gives at least (nd + 1) periodic solutions in a restricted stochastic model, while leaving full electrodynamics and physical applications unresolved.
A proof-based study establishes a unique, continuous extension under closedness and convexity, while an open half-disk example shows why the boundary matters.
The theoretical work gives a local upper limit on mean curvature, but not on the full second fundamental form in dimensions three and above.
The mathematical analysis gives the system a gradient-flow structure and establishes weak varifold-solution existence, with BV solutions only under additional conditions.
Best Prefix Selection scored bundles of skills for a language-model agent, but the evidence comes from one frozen-executor benchmark.
The Petz recovery map meets the proposed rules for reversing mathematical transformations, but its uniqueness remains an unproved conjecture.
An audit of 32 architecture–learning configurations found every forward-dynamics class represented, while global-gradient methods accounted for 16 entries.
A symbolic framework turns class-function evaluations into algebraic and combinatorial formulas, including a two-row rule for a set-mapping element.
An analytical framework separates local strut mechanics from changes in connectivity, backbone participation and load-path efficiency.
A mathematical proof shows that when a set with “nice recurrence” is divided into finitely many colour classes, at least one class retains the same property.
Random-wave backgrounds were associated with long-lived modeled structures, but the result does not establish real-world rogue-wave risk.