Latest papers
Markov Decision Processes (MDPs) are widely used as decision-making models, commonly specified over factored state spaces through state variables and their valuations. The exponential blowup in the number of states renders many reasoning…
We prove that the inclusion of left-invariant forms into the Dolbeault complex of a compact nilmanifold $M$ endowed with a left-invariant complex structure $J$ induces an isomorphism in cohomology in every bidegree, settling a long-standing…
We prove that every tower of connected quasi-torsor covers of a connected normal globally $F$-regular projective variety over an algebraically closed field of positive characteristic stabilizes, and that the Nori fundamental group scheme of…
Large language model agents often encounter related yet distinct tasks that share reusable solution patterns. Yet standard agentic reinforcement learning treats tasks as independent episodes, while existing approaches to skill learning…
This letter presents the first measurements of the average transverse momentum of prompt charged particles in $pp$ and $p\mathrm{Pb}$ collisions as a function of collision multiplicity and pseudorapidity. The data were recorded at…
The compressible Euler-Riesz equations arise in the modeling of a wide range of physical phenomena, including stellar dynamics, plasma physics, and mathematical biology. In this paper, we investigate the nonlinear stability and instability…
Precision tests of the Standard Model at low energy are increasingly limited by nuclear-structure theory rather than by experiment. We review two such cases: the two-photon-exchange correction to the Lamb shift in muonic atoms, and the…
The ability of large language models (LLMs) to process and generate text has introduced potential for applications in information extraction (IE). While it's debated whether LLMs outperform smaller fine-tuned models for classification…
Let $(X,\chi,k)$ be a triple consisting of a smooth, compact hyperbolic Riemann surface $X$ of genus $g$, and an $m$ dimensional unitary multiplier system $\chi$ of admissible weight $k$. Our first result establishes an analogue of the…
An interluminal interface, a space-time boundary propagating at a velocity between the group velocities of the surrounding media, enables extraordinary wave phenomena such as nonreciprocal amplification and analogues of Hawking radiation. A…
LLM-based code agents have advanced repository-level software development through iterative interaction with codebases and tools. However, feature development requires integrating new behaviors into existing architectures through coherent…
In this paper we prove that Dubois--Prade's eight possibilistic operators in Formal Concept Analysis arise canonically from Kan extensions of the underlying boolean profunctor. This provides a conceptual explanation for the result that…
Many kinds of data have structure along one or more axes: words in a sentence, pixels in an image, nodes in a tree, frames in audio, or cells in a 3D volume. Along one axis, order matters: "the dog bit the man" is different from "the man…
For a natural number $a$, let $E_{2a}$ be the Mordell elliptic curve $X^3 +Y^3=2a$. We give an explicit construction of (mock) Heegner point on the Mordell curve $E_{2p}$ for a prime $p\equiv 4 \mod 9$ and $E_{2p^2}$ for a prime $p\equiv 7…
Latent communication in large language model (LLM)-based multi-agent systems (MAS) transmits continuous internal representations instead of text, but greater representational capacity does not establish that the receiver uses task-relevant…
Many security proofs for quantum key distribution (QKD) require Bob's measurement to have a tensor-product structure across protocol rounds, with some techniques requiring the stronger independent-and-identically-distributed (IID)…
We demonstrate an end-to-end hybrid quantum-classical optimisation framework based on Benders decomposition, capable of solving mixed-integer linear programming (MILP) problems. The framework builds on a previously presented hybrid…
Temporal logic (TL) provides a compositional language for the formulation of long horizon robotic tasks, but existing TL-conditioned trajectory generators can sidestep perception-to-symbol binding by encoding exact object geometry in the…
Multimodal large language models increasingly use sketches, annotations, tools, and intermediate images during reasoning, but it remains unclear whether they truly rely on these visual states. Existing benchmarks are limited both by task…
This paper investigates the classification of spatio-temporal patterns generated by linear cellular automata with uniform weights (LCA-UW) over the ring ${\mathbb Z} / n {\mathbb Z}$. While these systems are governed by the state size $n$…