Papers

Transmission spectroscopy of exoplanets provides an important window into their atmospheric composition, structure and dynamics, especially in the JWST era. As the quality of the observational data and therefore the precision of our constraints improves, we become increasingly sensitive to sources of systematic bias in our models and observations. One significant example of this is the Transit Light Source Effect, in which heterogeneous features in the stellar photosphere - spots and faculae - imprint additional spectral signatures onto exoplanet transmission spectra. The effects of this stellar contamination must be accounted for or removed in order to accurately recover the planet's atmospheric properties. Including parameterized models of spots and faculae within spectral retrieval analysis is increasingly adopted as a solution to this problem, but is limited by the accuracy of models of stellar spectra. In this paper, we outline an observational strategy combining simultaneous data from JWST and the Hubble Space Telescope that would enable us to further constrain these models. We perform synthetic retrievals to demonstrate how we could constrain starspot and faculae parameters; we test the impact of using different model spectra in the retrieval; and we consider the impact of variable stellar activity on coadding transits to achieve the desired signal-to-noise. We find that ultraviolet and optical wavelengths are key for breaking degeneracies between stellar contamination and planetary parameters.

Tornadoes pose substantial risk to human life and property in the United States, causing more than 50 fatalities and \$100 million of property damage on average annually. When tornadoes are likely, weather radar provides critical information for forecasters by providing information on storm morphology, storm motion, and intensity trends. Additional tools such as satellite and numerical weather prediction model runs can provide useful short-term information for understanding changes in storm characteristics. This work demonstrates a U-Net deep-learning system for nowcasting the evolution of radar reflectivity following tornadogenesis, which can provide value to forecasters by synthesizing large amounts of input data (e.g., radar imagery, near-storm environment data) and generating predictions of radar reflectivity from its inputs. Inputs to the model are radar imagery from the Multi-Radar Multi-Sensor (MRMS) dataset and near-storm environment data from the High-Resolution Rapid Refresh (HRRR) numerical weather prediction model. The U-Net is trained on a dataset of tornadic storms to produce 30 minutes of probabilistic predictions of radar reflectivity following tornadogenesis, with probabilistic predictions obtained by predicting parameters of the SinhArcSinh, or SHASH, distribution. The model produces physically realistic predictions of radar evolution, achieves comparable skill to next-hour forecasts from the HRRR, demonstrates reasonable probabilistic calibration and is accompanied by a variety of explainability methods to improve understanding by end users. Additionally, predictions from the model can be obtained much more quickly than those from a numerical weather prediction model. With further development, this model could be extended to nowcast radar reflectivity evolution in an operational setting.

Of the approximately 4,500 Oracle Bone Inscription (OBI) characters discovered from the Shang dynasty, only about 1,600 have been deciphered. Many computational approaches compare OBI with glyphs from one historical period at a time. However, during the evolution of Chinese characters, significant structural or semantic changes often occur in uncertain dynasties. A single-period reference may be insufficient when relevant forms change substantially between observed eras. Therefore, we propose the Manifold-based Script Evolution Framework (MSEF), a framework that models the evolution series (OBI, Bronze, Seal, Clerical, Regular) of Chinese characters as the continual evolution of a manifold space. MSEF represents each character as an era-specific manifold point and learns continuous inter-era transition rules via Neural Ordinary Differential Equations. Both manifold space and transition dynamics can be trained end-to-end through character evolution pairs across any two eras.

Tool-based image editing (image retouching) is commonly formulated with autoregressive multimodal large language models (MLLMs) that sequentially generate reasoning, tool selections, and parameter values. In this work, we present a novel approach to tool-based image editing by framing the task as a flow matching problem. We introduce FlowTool, a framework that directly models the distribution of high-quality tool parameters conditioned on the input image and user instruction using conditional rectified flow. FlowTool combines a vision-language model backbone for multimodal understanding with a Diffusion Transformer parameter generator that transforms Gaussian noise into an editing plan. We train FlowTool with a two-stage supervised flow-matching curriculum, followed by reward-based post-training. Across MMArt-Bench, FlowTool-Eval, ArtEdit-Bench, and MIT-Adobe5K, FlowTool achieves significantly stronger reference-based performance than specialized MLLM editing agents and proprietary MLLMs, while remaining competitive with proprietary models under reference-free evaluation. Moreover, FlowTool significantly improves inference efficiency, reducing latency by at least $50\times$ while requiring nearly $2\times$ less memory than the compared baselines. These results demonstrate that tool-based image editing can be effectively modeled as conditional generation over structured continuous editing parameters, without autoregressive reasoning.

Music producers search libraries of isolated instrument recordings, called stems, for sounds resembling parts of an existing song. Neural retrieval systems address this by mapping audio to embeddings and ranking library stems by their similarity to the query. The leading method, Contrastive Instrument Retrieval (CIR), encodes the mixture as a single embedding, but it works best when a user specifies the target's instrument family. We introduce Stembed, which encodes a mixture as several slot embeddings representing candidate stems. During training, we construct mixtures from stems of the same song and match their slot embeddings to those of the isolated stems. The slot embeddings from mixtures inherit the stem identities of their assigned solo embedding, enabling a contrastive loss. On mixtures from held out MoisesDB artists, Stembed outperforms a CIR-style baseline when both search the full stem library. Even when predicting the stem count itself without family labels, Stembed exceeds the baseline's family-filtered R@1. Our website demonstrates how users can select a slot by inspecting the tags of its retrieved stems.

Mobile GUI agents operate through a perception–action loop: at each step they screenshot the device, invoke a vision–language model (VLM), and emit an action. It is slow, costly, and brittle, yet most of what it does is navigation—and everyday navigation is static, ordered, and endlessly repeated. We present PhoneCLI, which compiles an app's GUI navigation into callable commands, without any app-internal API, runtime instrumentation, or model training. Offline, PhoneCLI explores a target app from the outside and distills its screens, interactive elements, and navigation edges into a semantically annotated map; each screen yields one deterministic command: a replay sequence that reaches it. Online, the agent selects a command, verifies it before execution, and then executes it deterministically in sub-second time at zero VLM cost; open-ended interaction and every failure of the compiled path fall back to the embedded VLM interpreter, exactly the pure VLM agent, so compilation can only help. On AndroidLab, PhoneCLI improves the task success rate while reducing steps and token consumption, and it transfers to AndroidWorld's official M3A agent with consistent efficiency gains. What PhoneCLI compiles is the app's navigation rather than one run, so it serves new tasks, not only repeated ones.

We study the truthful and fair allocation of $m$ indivisible goods to $n$ strategic agents with additive valuations. Babaioff, Feige, and Manaker Morag (FOCS~2026) gave a randomized mechanism that uses only the agents' rankings of the goods, is truthful-in-expectation (TIE), and guarantees every agent $1/(H_{n-1}+2)=\Theta(1/\log n)$ of her maximin share (MMS) in every realized allocation, where $H_{n-1}$ is the $(n-1)$th harmonic number; this is nearly the best possible with rankings alone. They conjectured that cardinal information allows TIE mechanisms to achieve a constant ex-post MMS guarantee. We confirm this conjecture: our TIE mechanism gives every agent at least $1/6$ of her truncated proportional share (TPS), and hence of her MMS, in every realized allocation. Moreover, the mechanism is ex-ante envy-free and can be implemented in polynomial time as an explicit lottery over at most $m(n-1)+1$ allocations.

In this work, we establish a sharp logarithmic Choquard inequality by combining Beckner's entropy inequality with the sharp Pitt-type inequality. Motivated by this estimate, we classify the positive classical solutions, in a suitable integrability class, of the critical logarithmic Choquard problem \[ \mathcal{L}_\Delta u = \sigma u + \frac{1}{\|u\|_2^2} \left(G_{\ln}(u)-\frac{4}{N}\int_{\mathbb{R}^N}u^2\ln u\,dx\right)u \qquad\text{in }\mathbb{R}^N, \] where $\sigma\in\mathbb{R}$, $N\geq 1$, $\mathcal{L}_\Delta$ is the logarithmic Laplacian, and $G_{\ln}(u)=\ln(1/|x|^4)*u^2$. We construct explicit solutions as limits of critical fractional Choquard bubbles and prove that every positive classical solution in the prescribed class has the form \[ u_{\sigma,t}(x) = e^{\frac{N}{4}(\sigma-B_{N,\mathcal{L}})} B_{N,0}\left(\frac{t}{t^2+|x-x_0|^2}\right)^{N/2}, \qquad t>0,\quad x_0\in\mathbb{R}^N, \] where \[ B_{N,0} =\left(\frac{\Gamma(N)}{\pi^{N/2}\Gamma(N/2)}\right)^{1/2}, \qquad B_{N,\mathcal{L}} =2\ln 2+4\psi(N/2)-2\psi(N)+\frac{4}{N}\ln B_{N,0}. \]

We determine the optimal query complexity of ground-state preparation to trace-distance error $\varepsilon$ when an energy threshold in the spectral gap is known. Let $U_H$ be an $\alpha$-block-encoding of a Hamiltonian with unique ground state $|\psi_0\rangle$, and suppose $|\langle\psi_0|U_I|0\rangle|\ge\gamma$ for a state-preparation oracle $U_I$. The threshold lies at least $\Delta/2$ above the ground-state energy and at least $\Delta/2$ below every excited-state energy. We give two algorithms that prepare a state within trace distance $\varepsilon$ of the ground state. One uses $O((\alpha/\Delta)(\gamma^{-1}+\log(1/\varepsilon)))$ calls to $U_H$ in expectation; the other uses $O((\alpha/(\gamma\Delta))\log(1/\varepsilon))$ calls to $U_H$ in the worst case. We prove a lower bound matching the expected query count; the corresponding worst-case lower bound follows from Somma and de Wolf [SdW26]. The respective bounds on calls to $U_I$ are $O(1/\gamma)$ in expectation and $O(\gamma^{-1}\log(1/\varepsilon))$ in the worst case. On $(N+1)$-dimensional systems, these $U_I$ bounds are also optimal when the expected or worst-case count of $U_H$ calls, respectively, is $o((\alpha/\Delta)\sqrt N)$. Both algorithms use a constant-accuracy spectral filter to construct a purifier, which we then sequentially compose during amplitude amplification to prepare a state with constant overlap with the ground state. The expected-query algorithm repeats the preparation followed by one high-accuracy spectral filter until success. The worst-case algorithm uses filters of increasing accuracy and limits the total number of queries.

BaxZr1-xTiO3 ceramics with 0.15 < x < 0.25, which have high dielectric permittivity, small leakage currents and low dissipation factor, are promising materials for tunable capacitor devices, multilayered ceramic capacitors and piezoelectric actuators. These materials reveal relaxor properties due to the local chemical strains caused by the isovalent substitution of Ti4+ ions by Zr4+ ions with a larger ionic size. However, to the best of our knowledge, the pyroelectric properties of BaxZr1-xTiO3 ceramics with x~0.2 have not been studied. To fill the gap in knowledge, we perform the pyroelectric thermo-wave probing of BaZr0.2Ti0.8O3 ceramics prepared by the solid-state synthesis. Results of the thermo-wave probing of the most interesting polar states, namely polarized, depolarized, relaxed and restored polarized states, reveal the pronounced pyroelectric response that can be strongly asymmetric with respect to the opposite surfaces of the ceramics. The asymmetry and profiles of pyroelectric response depend significantly on the pre-history of the electric field cycling indicating possible non-ergodic relaxor-type polar states in the ceramic sample. X-ray diffraction spectrum, recorded at room temperature, reveals the virtual absence of the macroscopic tetragonality and the weak asymmetry of the (200) peak, which indicate the small tetragonality inside the Ti-enriched polar nanoregions. Decrease of the relative intensity of the Raman band at 714 cm-1 occurring upon heating evidences the diffuse ferroelectric-paraelectric phase transition between 30 C and 40 C. The temperature dependence of the dielectric permittivity obeys modified Curie-Weiss law with power 1.4, indicating that both ordered ferroelectric and relaxor states coexist in the BaTi0.8Zr0.2O3 ceramics.Results can be useful for elaboration of lead-free relaxor ferroelectric ceramics for advanced pyroelectric applications.

We study the sharp-interface limit of the fractional Allen–Cahn equation in $\R^n$, $n\geq 2$, in the strongly nonlocal regime $s\in(0,\frac12)$, for initial data consisting of finitely many nested transition layers. We identify a coupled geometric law for the motion of the resulting fronts: each velocity contains fractional mean curvature and a nonlocal interaction potential generated by the other fronts. These interactions persist while the fronts remain separated, in contrast to the independent motion in the critical regime $s=\frac12$. Assuming that the coupled law admits a smooth evolution with strictly nested sets on a given time interval, we prove that the Allen–Cahn solutions converge locally uniformly away from the moving fronts to the corresponding integer-valued phases.

Interior-point methods (IPMs) are among the most widely used algorithms for constrained optimization, yet their Newton-based search directions require costly second-order information and large linear-system solves. Learning to optimize offers cheaper updates learned from data, but the singular behavior of logarithmic barriers near constraint boundaries makes IPMs highly sensitive to perturbations, complicating both warm starting and learning reliable updates. We introduce pdLIP, an IPM for smooth nonlinear programs that integrates learned preconditioning with pdProj, an all-shifted primal-dual projected-search IPM. A shared coordinate-wise recurrent network predicts a positive diagonal preconditioner that scales the right-hand side of the reduced Newton system for the primal step, and the remaining slack and multiplier directions are recovered analytically. The learned iterations avoid Hessian evaluations and Newton-system solves, using only first-order and coordinate-wise operations amenable to GPU parallelization. Training is self-supervised, with a loss based on a penalty-barrier merit function and the residual of perturbed optimality conditions, requiring neither target directions nor precomputed solutions. Primal and dual shifts mitigate the barrier's sensitivity to perturbations near constraint boundaries, enabling effective warm starting. Across four classes of 200-dimensional convex and nonconvex constrained problems, pdLIP warm starts reduce pdProj refinement iterations by 63-67% compared with cold starts at the same KKT residual tolerance of $10^{-8}$, with negligible warm-start generation cost relative to the subsequent pdProj solve. Improvements persist on box-constrained QPs with 1000 variables and extend to applications including portfolio optimization, support vector machines, and a nonlinear control example.

Gated Linear Attention (GLA) Transformers advance linear recurrent models through data-dependent gating, but face a core limitation: the fixed-capacity memory matrices across all heads operate at a single temporal resolution, where each token is processed individually, forcing them to simultaneously encode local syntactic patterns and long-range semantic structure, creating a representational bottleneck that gating alone is insufficient to resolve. We introduce Multi-Scale Gated Linear Attention (MS-GLA), which addresses this by distributing attention heads across multiple temporal resolutions. Coarser resolutions pool longer token spans naturally specializing toward long-range dependencies, while finer head groups retain sensitivity to local syntactic structure. A learnable, input-dependent fusion layer dynamically recombines head group outputs at each timestep, expanding effective memory capacity without increasing per-head state size. This multi-resolution decomposition draws on principles from Multi-Scale State-Space Models (MS-SSM), adapting them to the gated linear attention setting. We evaluate MS-GLA on language modeling, recall-intensive tasks, and long-context generalization. Across all settings, MS-GLA consistently achieves higher accuracy and lower perplexity than GLA at matched parameter counts, with up to 18.9% improvement on recall-intensive tasks and 9.5% lower average perplexity on language modeling benchmarks, validating multi-temporal resolution decomposition as a principled and effective extension of Gated Linear Attention.

Large language models (LLMs) reliably perform entity copying, in which a model copies tokens referring to an entity, termed entity tokens, from the prompt into its output to answer a question. Although entity copying is straightforward for most LLMs, existing research does not provide a systematic account of which layers specialize in this fundamental task or how other tokens in the same sequence, termed context tokens, influence the model's ability to copy the entity tokens. To address these questions, we conduct experiments on Qwen3-8B using two novel methods: genie-in-a-bottle, which controls exactly which layers can participate in an entity-copying task, and attention lobotomy, which cuts off specific tokens' attention to entity tokens without affecting the remaining attention distribution. We find that two distinct groups of layers in the second half of the model are both necessary and sufficient for entity copying. Moreover, in addition to the decoding position's attention to entity tokens, context tokens' attention to entity tokens also proves necessary for copying the exact tokens, even though context tokens do not store entity information themselves unless they satisfy particular semantic properties. Our findings establish the critical role of late layers in entity copying under the guidance of context tokens, calling for future work on how models propagate and consume entity information.

The wakeup problem captures a fundamental symmetry-breaking challenge among devices sharing a communication channel. We study the dynamic setting, where packets become active at arbitrary times on a time-slotted multiple access channel. In each slot, a transmission succeeds if and only if exactly one packet transmits; two or more simultaneous transmissions cause a collision. The goal is to obtain a successful transmission quickly. Prior work on wakeup has largely focused on the number of slots until the first success, referred to as the latency. However, a collision may incur substantial additional delay, represented by a per-collision cost $C$. We therefore seek to control both latency and the collision cost of an execution, defined as $C$ times its number of collisions. We design and analyze a randomized algorithm for dynamic wakeup, Lowball, without collision detection or knowledge of the number of packets, $n$. Fix a constant $0<\epsilon\le 1/2$. There is a constant $K>0$ such that, when $C\ge K\lg^{1/\epsilon} n$, Lowball has expected latency $O(C^{1/2+\epsilon}\ln C)$ and expected collision cost $O(\sqrt{C})$. Below this threshold, both expectations are $O(n\log^{\Theta(1/\epsilon)} n)$. These guarantees hold against an adaptive, non-anticipating adversary, and the algorithm succeeds with probability 1. For algorithms in which each packet's transmission probability depends only on $C$ and the packet's local age, with packets activated together using the same probability schedule, we prove that the maximum of expected latency and expected collision cost is $\Omega(\sqrt{C})$.

Solving large ensembles of small, independent ordinary differential equations (ODEs) is a common task in computational science and engineering, arising for example in Bayesian parameter estimation, Monte Carlo uncertainty quantification, and the integration of uncoupled physical systems such as the linearised Einstein-Boltzmann equations of cosmology. Modern graphics processing units (GPUs) with thousands of cores are well suited to such parallel workloads, but realising this potential requires careful attention to GPU-specific architectural constraints. We present modax, a Python library of GPU-accelerated ODE solvers specifically optimised for solving large ensembles of low-dimensional, independent problems. Our solvers are compatible with the JAX ecosystem but are written in a low-level, thread-based programming paradigm better suited to ensembles of highly divergent trajectories than their pure-JAX counterparts. Benchmarks on the Lorenz, Robertson and van der Pol lattice systems compare modax against diffrax, DiffEqGPU$.$jl and torchdiffeq across ensemble size, dimensionality, and trajectory divergence. Our explicit Tsit5 solver achieves 30x speed-up over diffrax on Lorenz systems, and our implicit Rodas5P solver achieves over 700x speed-up on Robertson systems, demonstrating the superiority of linearly implicit Rosenbrock-Wanner methods on GPUs. Our implicit solver scales well to higher dimensions thanks to the use of sparsity-exploiting techniques and data structures. We conclude by applying modax to three examples from the field of cosmology - Bayesian parameter estimation from primordial abundances, Monte Carlo uncertainty quantification of the global 21cm signal, and the computation of uncoupled Fourier modes in the primordial power spectrum - and demonstrate significant efficiency improvements in our ability to model physical phenomena described by low-dimensional (<200D) ODEs.

In the quantum linear systems problem (QLSP), we are given query access to a $d$-sparse $N\times N$ matrix $A$ with condition number $\kappa$, and the ability to prepare a quantum state proportional to a vector $\vec b$. The goal is to output an $\epsilon$-approximation to the quantum state proportional to the solution $\vec x$ of $A\vec{x}=\vec{b}$. Following a long line of work, the best previously known quantum algorithms for the QLSP had query complexities $O(\kappa d\log(1/\epsilon))$ and $\kappa\sqrt d(\kappa d/\epsilon)^{o(1)}$, while the best known lower bounds were $\Omega(\kappa\log(1/\epsilon))$ and $\Omega(\kappa\sqrt d)$. We improve these bounds and show that the complexity of the QLSP is $\Theta(\kappa\sqrt d\log(1/\epsilon))$. We also resolve an open problem of Berry and Childs by showing that any $N\times N$ unitary can be implemented with bounded error using $O(\sqrt N)$ queries to its matrix entries.

AI agents run shell commands, edit repositories, send email and move money with their operators' credentials, yet the record of what they did is usually a log the agent, or whoever runs it, can edit. We present Tracekit (open source, MIT, version 1.4.1), which turns each agent run into evidence a third party can check offline. A signer that the agent cannot reach outside dev mode decides every tool call against its own policy before it runs, and signs the call, decision and result as Ed25519 records in a per-run hash chain inside an RFC 6962 Merkle tree whose checkpoints independent witnesses can cosign. A run exports as one .tkb bundle that two independent verifiers (Python, and TypeScript, also as a browser page) check against keys the verifier pins, reporting integrity and assurance separately. Disagreements between capture layers become signed records. Tracekit also records the person's request and what the model said before each call, holds calls that break a limit the request stated, compares what was asked, said and done, and finds which input caused an action by replay. Of 21 experiments: a signed log alone detects 6 of 8 tamper classes and a witness the other two; every insider attack in two suites is caught; the signer round trip is 0.45 ms median; no acknowledged record was lost in 60 kill -9 rounds; reconciliation flags 12 of 12 seeded discrepancies; the verifiers reject all 33 forgeries. Policy rules are a tripwire, not the boundary: they stop 570 of 570 harmful calls in our corpus but 211 of 420 (50.2%) written blind to them; the protection is a signer and credentials the agent cannot reach. On a held-out pool, a failed test claimed as passing is found in 86% of sessions with 5 false findings in 1,812 honest calls, and replay names the planted cause in 6 of 6 confirmed harmful runs with a 7B model.

Real-world capture is heterogeneous: perspective, fisheye, and $360^\circ$ panoramic images can coexist within a single reconstruction task, yet most feed-forward 3D reconstruction models assume perspective imagery and a uniform input representation. Recent models handling several camera types are either informed of the camera type for each view or reconstruct one image pair at a time. No single-pass method reconstructs mixed-camera tuples containing full panoramas from images alone. We present MEOW, a feed-forward system that jointly reconstructs metric pointmaps and camera poses from one N-view tuple mixing perspective, fisheye and full-panorama images, in a single forward pass from images alone: no calibration, distortion parameters, camera-type labels or poses are supplied for any view. Our guiding design philosophy is to treat heterogeneous-camera reconstruction as a data-adaptation problem rather than an architectural redesign. MEOW retains a perspective-pretrained backbone and learns heterogeneous cameras entirely from a procedural data engine, which renders each scene across a continuous manifold of camera models with exact rays and depth, and certifies covisibility for every camera-sampled training tuple. Trained on synthetic tuples only, MEOW transfers zero-shot to real captures: on heterogeneous 2D3DS tuples it achieves 79.9 mAA@30 against 53.8 for Wid3R given the camera type of every view; on our laser-scanned mixed-camera benchmark it registers every four-view mixed tuple with 79.4 AUC@30. The data engine, benchmark, and complete evaluation pipeline will be released.

Population-based optimization methods often use previous search information through successful solutions, parameter adaptation, or operator performance, but they rarely retain the context in which a search behavior succeeded or failed. We introduce Cognitive Memory-Driven Optimization (CMDO), a derivative-free population-based optimizer that represents experience as the relationship between search context, search behavior, and observed outcome. CMDO organizes these experiences across working, episodic, and consolidated memory, retrieves them according to similarity with the current search state, and uses both positive and negative evidence to guide subsequent search. Retrieved experience does not replay previous candidate locations; instead, it selects search recipes that are reconstructed from the current population through exploratory, directed, and local search behaviors with adaptive search geometry. We evaluate CMDO on selected Blackbox Optimization Benchmarking test suite on COCO (BBOB/COCO) and Congress on Evolutionary Computation 2017 (CEC2017) problems against DE, CMA-ES, SHADE, GWO, HHO, and ORCA, and further study its application to seven-parameter photovoltaic model estimation using measured current–voltage data. The results show problem-dependent but competitive optimization performance, including the lowest median error among the compared methods on CEC2017 F10. More importantly, analysis of the search traces shows that context-dependent recall changes the distribution of executed search behaviors, while unsuccessful experiences remain available as negative evidence for later decisions, showing that accumulated experience directly influences subsequent search behavior. These results support the use of explicit context–behavior–outcome memory as an active mechanism for controlling population-based search.

In the quest to identify further quantum algorithms exhibiting superpolynomial speed-ups, a recurring theme is that the complexity of a problem is largely shaped by the input access model. Here, we study this phenomenon for the state hidden subgroup problem (StateHSP), a quantum generalization of the hidden subgroup problem in which the goal is to identify the symmetries of an unknown quantum state. For finite abelian groups, existing Fourier-sampling algorithms use $O(\log(|G|)/\epsilon)$ copies of the state, but whether this scaling is optimal has remained open. We settle the complexity of the abelian StateHSP in both the previously studied sample model and a new query model, which is a stronger and operationally natural generalization that provides access to the state-preparation unitary and its inverse. In the query model, we give a time-efficient quantum algorithm using $O(\log(|G/H|)/\sqrt{\epsilon})$ forward and inverse queries, and prove a matching $\Omega(\log(|G/H|)/\sqrt{\epsilon})$ lower bound which holds even in the stronger conjugate-query and controlled-query settings. By contrast, we show that in the sample model, $\Theta(\log(|G/H|)/\epsilon)$ copies are both sufficient and information-theoretically necessary, even if one allows for arbitrary collective measurements. Thus, the quadratic improvement in $\epsilon$ genuinely arises from coherent access to the preparation circuit. As applications, we obtain faster algorithms for learning stabilizer groups, locating unentanglement, and identifying hidden translation symmetries.

We study a quantum battery consisting of $N$ qubits interacting uniformly with each other. Starting from its ground state, the battery is charged by introducing periodic kicks in the transverse direction along with adding disorder in the interaction direction. Although the battery Hamiltonian along with the initial state preserves the permutation symmetry, the dynamics in presence of disorder during the charging phase takes the system out of the permutation symmetric subspace, resulting to a larger amount of stored energy as compared to that without disorder. In the limit of large disorder, this interplay brings in a possibility of a chaotic charging phase when stored energy can be exactly calculated using random matrix theory. More importantly, we also find that the energy fluctuations decrease and become negligible in the limit of large disorder, which is also when the energy stored is maximum, a condition highly desirable for the optimum functioning of the battery. We complete the battery cycle by considering the storage phase in the presence of dephasing noise. We find that for larger disorder strengths, the ergotropy remains close to the stored energy, indicating nearly complete energy extractability with an efficiency approaching unity.

Artificial intelligence models deployed as clinical decision support tools often suffer substantial performance degradation over time and across institutions due to covariate shift, concept drift, and cross-system heterogeneity. Fully retraining complex models is frequently infeasible, particularly in EHR settings where labeled outcome data are scarce and regulatory constraints limit model modification. We propose GUARD (Guided and Uncertainty-Aware Robustness to Domain shift), a unified statistical framework for updating an existing deployed model using limited labeled target data and multiple heterogeneous source populations while guarding against future distributional shifts. GUARD formulates robust multi-source transfer learning as a semi-supervised, adversarial optimization problem anchored at the current model, and employs double machine learning with cross-fitting to obtain debiased, efficient estimates that leverage abundant unlabeled target covariates. It then constructs an uncertainty-aware, guided group DRO estimator that combines source and target information while accounting for sampling variability in the source models. Our framework jointly provides principled robustness to domain shift, efficient semi-supervised estimation, and valid statistical inference for multi-source transfer of clinical prediction models. We demonstrate the utility of GUARD via extensive simulation experiments and a real world application to predicting future disease activity in rheumatoid arthritis from EHR data spanning more than a decade. GUARD recalibrates a pretrained model with only a small number of labeled records per year and remains accurate over multi-year horizons where target-only and standard transfer estimators degrade sharply.

We develop a new framework to perform a virtuality-resolved description of jet propagation in heavy-ion collisions using perturbative techniques. The role of virtuality is rigorously identified by first deriving a factorized jet cross section that defines an in-medium jet function. This new definition allows to extend the usual BDMPS-Z formalism to include the virtuality-differential transverse momentum broadening distribution and medium-induced soft gluon spectrum. We study both the cases of a medium which is created with some delay time $\tau_0$ after the hard collision as well as the case in which $\tau_0\to 0$. Integration over the complete virtuality range recovers the standard BDMPS-Z results, while finite virtuality reveals the interplay between vacuum-like evolution and medium-induced dynamics. We identify the ratio between the medium formation time and the jet formation time as the variable controlling which initial virtualities contribute appreciably to the jet function.

Mobile manipulation extends robot interaction beyond a fixed kinematic workspace by making the reachable region itself controllable. This flexibility introduces two central challenges: spatially grounded perception under continuous ego-motion and coordinated control of heterogeneous arm and base actions. Existing approaches strengthen geometry through explicit 3D representations or predictive world models, and often decouple mobility and manipulation into separate action streams. We argue that effective mobile manipulation requires not only decoupling, but also representations that support efficient cross-stream collaboration. We present MM-ABC, a foundation model built around Seeing, Coordinating, and Imagining Arm-Base Collaboration. MM-ABC combines sparse multi-level VLM features for spatial perception; a training-only future branch that uses world imagination and geometric intent as extra supervision, strengthening perception and manipulation-intent prediction and improving the overall learning signal; and MM-APT, which coordinates separate manipulation and mobility streams through masked joint attention and clean-action x-prediction. In controlled ablations, replacing clean-action prediction with velocity prediction lowers success on RoboCasa365 composite-seen tasks from 32.8% to 29.2%, and removing future supervision or multilevel conditioning causes larger drops. We pretrain MM-ABC on 5,000+ hours of heterogeneous robot data spanning 400K+ episodes, 12 datasets, and 17 embodiments. Experiments cover EBench, RoboCasa365, ManiSkill-HAB, LIBERO, LIBERO-Plus, and real-world mobile manipulation. MM-ABC achieves 44.71% success on EBench, 61.2% on RoboCasa365, 99.1% on LIBERO, 82.8% on LIBERO-Plus without perturbation training, and 83% mean success on five real-world tasks.

Multi-robot trajectory planning is a fundamental problem in multi-robot coordination but remains computationally challenging due to its nonconvex, multimodal, and high-dimensional nature. This work builds upon D4orm, a dynamics-aware diffusion-denoising framework, and develops a family of planning architectures for diverse operational requirements. Unlike conventional numerical optimization methods, D4orm employs sampling-based optimization to generate solution trajectories through massively parallel sampling, leveraging modern computing architectures such as GPUs. Its diffusion-denoising structure iteratively optimizes deformations to candidate control trajectories, providing an efficient and versatile paradigm for generating kinodynamically feasible and conflict-free trajectories. Using D4orm as the building block for advanced planners, we present a decoupled planner for improved scalability, an online receding-horizon planner with feedback control, and a distributed planner for resource-constrained settings. Evaluations with differential-drive and holonomic robots in 2D and 3D environments demonstrate that D4orm-based approaches find high-quality solutions faster and more reliably than other sampling-based optimization methods, such as MPPI, as well as a learned diffusion-model-based method. We further demonstrate zero-shot deployment on ten real quadrotors with obstacles, large-scale deconfliction with 100 simulated robots, and fully onboard distributed `lifelong' operation with six ground robots. Overall, these results establish diffusion denoising as a scalable and reliable framework for multi-robot coordination. Code and video: https://github.com/proroklab/d4orm

The LUX-ZEPLIN (LZ) experiment has reported a single nuclear-recoil candidate at $E_R \simeq 248$keV. LZ interprets it in terms of inelastic dark matter that scatters endothermically from its ground state, $\chi_1 N \to \chi_2 N$. In pseudo-Dirac models the excited state $\chi_2$ can survive to the present day, and exothermic scattering, $\chi_2 N \to \chi_1 N$, is then also open. The published LZ intervals do not describe this case. We rebuild the LZ statistical inference and validate it against the published intervals. We then apply it to a halo that contains a fraction $f$ of $\chi_2$. For $f \gtrsim 10^{-5}$–$10^{-2}$, depending on the mass splitting, the exothermic contribution dominates. For $f=1/2$ the cross section required by the candidate drops by factors of $80$ to $1.6\times10^4$ for $\delta = 250$–$350$~keV, and the kinematic cutoff of the endothermic signal disappears. A benchmark that explains the event with $\chi_1$ alone lies a factor of $30$ above the upper limit when half of the halo is $\chi_2$.

Tandem mass spectrum prediction supports compound identification across metabolomics, natural-product discovery, and environmental analysis. However, pretrained predictors often degrade under shifts in chemical space and acquisition conditions, while retraining domain-specific models from scratch is costly. We introduce SPARC, a retrieval-guided test-time specialization framework that adapts a pretrained predictor using a spectral reference library without accessing test-query spectra. For each target query, SPARC retrieves chemically related reference spectra to recalibrate fragment intensities within the learned fragmentation space. During Transfer, SPARC combines reference-guided spectral adaptation with reliability-aware consistency, using reconstruction behavior on retrieved spectra to selectively preserve trustworthy predictions during continual specialization. Across MassSpecGym, NPLIB1 and application-specific GNPS libraries, SPARC improves spectral prediction under multiple transfer settings. These results establish retrieval-guided test-time specialization as a practical strategy for extending pretrained MS/MS predictors to specific chemical and acquisition domains, with continual test-time training providing further refinement during deployment.

We study the problem of learning a sparse quantum state, an $n$-qubit quantum state whose density matrix has at most $s$ nonzero matrix entries in an unknown product basis. While such states admit compact classical descriptions, they can carry long-range entanglement that prevents reconstruction from local reduced density matrices alone. Therefore, previous learning approaches addressed such long-range-entangled states using many entangling gates to extract the necessary information. In this work, we show that sparse states can nevertheless be efficiently learned using only single-qubit measurements. Specifically, when the sparsity $s$ is constant, our algorithm can learn sparse states from single-qubit measurements with polynomial sample complexity and classical computational complexity. When $s$ grows polynomially with $n$, sparse states can still be learned from single-qubit measurements with polynomial sample complexity, although efficient classical computation is not guaranteed in general. In this regime, however, the classical computational complexity becomes quasipolynomial when the state is sparse in an unknown basis that is a product of a known fixed finite set of single-qubit bases (e.g., eigenbases of Pauli operators). These results establish efficient learning of sparse states with long-range entanglement without entangling gates, and the single-qubit measurement requirements make our algorithms compatible with current quantum devices.

While structure-preserving methods for the forward simulation of Hamiltonian systems are well established, the mathematical foundations of their data assimilation and filtering counterparts remain comparatively less developed, particularly regarding rigorous accuracy guarantees. To address this gap, we propose an online filtering algorithm to reconstruct an unknown function $u^\dagger$ from finitely many measurements, assuming that $u^\dagger$ solves a parametric Hamiltonian PDE with unknown inputs. The method uses low-dimensional symplectic approximation spaces that evolve in time according to the PDE model and informed by the measurement data. The method can also be coupled with a dynamical sensor-placement strategy designed to optimize reconstruction stability. We derive error bounds in terms of stability and best approximation, and show that the reconstruction remains in a symplectic space and preserves energy up to the approximation error. Numerical experiments demonstrate the effectiveness of the method and the benefits of dynamic over static sensor placement.