Papers

A central challenge in molecular modeling is reconciling simulations with experimental observables, as force-field inaccuracies can distort equilibrium populations and long-timescale kinetics. While time-dependent restraints can improve simulated kinetics, time-dependent structural observables remain limited. Thus, most experimental data are time-averaged, informing equilibrium ensembles but not directly dynamics. Although ensemble refinement can improve thermodynamic accuracy, incorporating time-averaged observables into kinetic models remains difficult. Here, we introduce BICePs-reweighted Reversible DeepMSMs, combining Bayesian Inference of Conformational Populations (BICePs), variational learning of Markov processes, and maximum entropy (MaxEnt)/maximum caliber (MaxCal) principles to infer consistent thermodynamics and minimally perturbed kinetics. At its core is a reversible DeepMSM prior, obtained by applying an orthogonal transformation post-hoc to a pre-trained dynamics model (e.g., VAMPnet). This preserves the eigenspectrum exactly while yielding a valid transition matrix that is nonnegative, row-stochastic, and reversible. The reweighted model refines stationary populations, transition dynamics, and state-conditioned configurational landing densities. We validate the approach on a quadruple-well toy system and alanine dipeptide using backbone dihedral angles and J-coupling constants. In both systems, perturbing the observables induces predictable changes in the equilibrium ensemble and corresponding inferred kinetics. The resulting relaxation timescales and dynamical modes closely agree with analytical and MaxCal reference models. Furthermore, a diffusion-based generative model trained on MaxEnt landing densities produces physically realistic alanine dipeptide trajectories that reproduce thermodynamics and kinetics while satisfying the imposed experimental restraints.

We present CoBranchMR, a mixed reality (MR) system that enables distributed collaborators to work in parallel from different locations on the same digital representation of a physical object. CoBranchMR lets users branch an object into editable virtual copies, customize them independently, and then merge their work back into a shared object. When merging copies, the system displays potential conflicts on the object's surface and provides several resolution options. By adopting branch-and-merge workflows for embodied spatial collaboration, CoBranchMR introduces a new collaborative interaction model that supports parallel design, conflict resolution, and negotiation in remote creative work.

AI virtual cells aim to predict cellular responses to specified interventions, yet held-out predictive performance alone does not establish use of the supplied perturbation information. This prediction-claim gap matters in agentic model discovery, where language-model agents generate and revise predictors using score-based feedback. We introduce CELLAUDIT, which audits input-use claims by asking whether an input can enter the cited computation, whether fitted predictions depend on it, and whether that dependence improves prediction of observed response. On a paired morphology-transcriptomics perturbation benchmark (BBBC047), an agent-selected predictor attains a mean held-out Global Pearson correlation coefficient (PCC) of 0.3153 but remains invariant to compound replacement; a control-profile-only predictor reaches 0.3142. Source inspection identifies a compound-query pathway blocked by singleton key-value attention, and the invariance persists after refitting with disjoint control wells. In a stratified audit of 48 candidates across two linked tasks, 47 change predictions under compound replacement on both held-out folds, but only 20 show target-loss gains with intervals above zero on both folds. On BBBC047, falsification-guided revisions recover positive mean compound contributions while retaining gains over the control-profile-only baseline. In matched sci-Plex searches, audit-enriched feedback yields higher held-out performance and larger mean compound and dose contributions across five trajectories, although paired intervals span zero. Refitting fixed designs on an independently acquired cohort shows predictive generalization need not imply generalization of input-use claims: dose contribution persists, whereas support for compound identity does not. CELLAUDIT adds a falsification layer to agentic model discovery, moving from generate-score-revise toward discover-falsify-revise.

Transformer language models process sequences token by token in an autoregressive manner, making growing contexts increasingly expensive. Yet many adjacent token spans are highly predictable or frequently occur as stable units, suggesting that their representations may be compressible. We introduce a method for distilling sequential computation by replacing spans of input tokens with collapsed representations, computed on the fly by a lightweight merge module. This module generates a single surrogate embedding from a sequence of static token embeddings that captures the functional role of the multiple tokens, allowing pretrained models to operate on compressed inputs without architectural changes or re-training. We apply this approach during inference to compress both prompts and intermediate decoding steps, using a rollback mechanism to substitute stored multi-token KV cache entries with their single-step surrogates. Experiments across diverse models show that the merge module can be used to reduce effective sequence length by up to 40% with minimal accuracy degradation across language modeling evaluations and downstream tasks, including question answering, summarization, commonsense reasoning, and long-form mathematical reasoning. Additional lightweight adaptation of the merge module further improves the accuracy-compression trade-off in selected settings. These results demonstrate that sequential token computation in Transformers can be effectively approximated through condensed surrogate representations that approximate the original behavior without model updating.

Scaling laws have guided large-model development in computer vision and natural language processing, but the relationships among data, model size, and compute remain unclear for functional magnetic resonance imaging (fMRI) foundation models. Here, we conduct a controlled empirical study using pretraining data from more than 200 source datasets and over 10,000 GPU-hours of experiments. Holding the pretraining framework and downstream protocol fixed, we vary pretraining data size, model size, and training duration. Downstream performance generally improves with compute, yet models using similar compute can perform substantially differently. Additional pretraining data bring larger gains at larger model sizes, suggesting that data and model size should be scaled together. At matched compute, increasing pretraining data benefits more tasks than increasing model size, although the pattern varies across tasks. We then use in-distribution (ID) downstream performance to select the combination of pretraining data size, model size, and training duration at two fixed compute budgets. The resulting models are locked before out-of-distribution (OOD) evaluation. They achieve the highest average performance across the evaluated OOD tasks among the compared fMRI foundation models while using less pretraining compute. Overall, our results show that compute alone does not characterize fMRI scaling: performance depends on how pretraining data, model size, and training duration are combined.

When designing algorithms for geometric graphs, exploiting structural parameters can lead to significantly improved bounds. Two prominent parameters in this context are $c$-packedness and $\lambda$-low density, both of which locally restrict graph complexity. Parameterized algorithms based on these parameters have been developed for computing well-separated pair decompositions, balanced separators, as well as distance oracles. Nevertheless the practical applicability of algorithms parameterized by $c$ or $\lambda$ remains unclear. While $c$-packed and $\lambda$-low-density graphs have been proposed as realistic models for road networks, the actual parameter values of large real-world instances have so far remained unknown, and existing theoretical guarantees are partially too loose for practical usage. In this paper we first devise scalable implementations for the approximate computation of $c$ and the exact computation of $\lambda$. Our experiments on road networks with millions of edges reveals a significant gap between the two parameters. On the theoretical side we prove that $c\in O(\lambda\sqrt n)$ which complements the known result that $\lambda\in O(c)$. Furthermore we present improved parameterized algorithms for balanced separator computation that reduce the separator size in theory and practice. We also show how to compute a tree decomposition with a width linear in the respective parameterized balanced separator size in polynomial time. This structural result yields a variety of new algorithmic consequences. Among them is an exact distance oracle with query time $O(c)$ for $c$-packed graphs after polynomial-time preprocessing, which improves upon the previous $O(c\log n)$ bound. Our experiments show that the proposed techniques efficiently produce small balanced separators and enable the construction of concise exact distance oracles on large road networks.

Voice conversion (VC) systems fall into two families: non-parametric embedding-space methods, which need no trained model but degrade on short target utterances, and spectrogram-based neural architectures, which achieve strong quality via multi-module pipelines with tens of millions of parameters. We propose kNN-FM-VC, a single conditional flow-matching network that learns to approximate the kNN-VC mapping between WavLM embedding distributions of source and target speakers, replacing explicit pointwise kNN matching with a neural regressor trained on kNN-generated pairs. The model is conditioned on the target speaker via cross-attention and FiLM, and trained under three Gaussian conditional paths (Schrödinger bridge, straight line, and constant-variance Gaussian tube), enabling few-step sampling. Unlike Phoneme Hallucinator, which uses an upsampling stage followed by kNN matching, our 13M-parameter model performs conversion with a single learned network and supports one-step inference. On LibriSpeech, the one-step Gaussian Bridge achieves lower WER and higher estimated speech quality than FreeVC and Phoneme Hallucinator. Relative to kNN and kDOT, it substantially reduces WER.

Reliable power grid control requires sequential decisions under transmission constraints, time-varying demand, renewable variability, and imperfect load information. We investigate how load information quality affects control in a customized transmission network simulator inspired by Grid2Op and based on a DC power flow model. The environment includes stochastic spatial loads, temporally correlated renewable-like generation, energy storage, line protection, generator dispatch, and line reconnection. An Ensemble Score Filter (EnSF) corrects noisy load information before it is passed to the controller. A heuristic study first isolates the effect of this correction under a fixed feedback rule. Proximal policy optimization (PPO) agents are then trained for sequential generator and reconnection control, frozen, and evaluated with Forward, EnSF-corrected, and Truth load information on paired stochastic scenarios. EnSF consistently lowers load estimation error. Under heuristic control, the correction extends survival and reduces warning and overload exposure. With frozen PPO policies, it increases cumulative return and keeps performance closer to the Truth benchmark, with a larger advantage when load uncertainty is amplified beyond the nominal training distribution. In both experiments, EnSF reduces load-estimation error and improves the resulting control performance in the simplified transmission network.

The Plateau–Rayleigh instability of liquid jets governs a wide range of natural and industrial processes. While the classical linear stability theory accurately predicts disturbance growth rates and dominant wavelengths, it cannot quantify the amplitude and frequency spectrum of nanoscale initial surface disturbances. This critical missing information prevents accurate forecasting of breakup lengths and droplet generation. Conventional optical imaging is constrained by the diffraction limit and lacks the capability to resolve such nanoscale surface fluctuations. Here we introduce a Mie-scattering-based measurement technique to characterize nanoscale surface disturbances on liquid jets. The method employs a thin laser sheet to illuminate the jet, with scattered light captured at designated azimuthal observation angles. Based on Lorenz–Mie theory, we develop an inversion algorithm incorporating cosine similarity and cross-correlation to reconstruct jet surface disturbances from measured scattering fringes, achieving a sub-nanometer resolution for detecting diameter variations, with an overall measurement precision limited by a background noise floor of 0.18 nm RMS. Comparative analyses of forward and rainbow scattering signals further confirm that the surface disturbances become axially symmetric as the amplitude reaches a few nanometers, and thus satisfies the assumptions of the scattering model. Synchronous validations against high–speed imaging and scanning electron microscopy corroborate the quantitative accuracy of our scattering method. Applied in the near-nozzle region, the technique identifies two intrinsic signatures of initial perturbations: stochastic pulse-like fluctuations and low-frequency oscillations.

Objective assessment of robotic surgery uses instrument kinematics, which must be reconstructed when only video is available. We introduce a kinematic reconstruction network for estimating instrument position, orientation and jaw angle from monocular video. Our visual representation combines global attention pooling of frozen DINOv3 features with local pooling at instrument landmarks from fine-tuned SAM 3.1 masks. Our shared Transformer encoder and temporal convolutional heads integrate this representation with mask geometry, monocular depth and visual state estimates from arm-specific multilayer regression networks. Our position branch predicts displacement magnitude and direction separately to preserve traveled distance. We fit trajectories to predicted state observations and motion increments by differentiable weighted least squares, expressing quaternion observations relative to cumulative predicted rotations to obtain a quadratic orientation objective. We evaluate reconstruction across 2,802 Open-H episodes. Compared with LiveMAE on the main Open-H benchmark, our method reduces path-length mean absolute error from 0.45 to 0.34\,cm and increases temporal mean average precision for motion segmentation from 44.54% to 54.44%.

Movable antenna (MA) systems reconfigure the wireless channel by repositioning the antenna within a confined region. The position optimization behind this capability has been studied under the premise that the channel at every candidate position is known. However, acquiring that knowledge is costly: a position can be measured only after the antenna has moved there, which takes many slots at the speed of the actuator, and the channel drifts meanwhile. We formulate MA positioning as a non-stationary kernelized bandit with a reachability constraint, in which exploration, tracking, and the reward lost in transit are coupled through a single physical action. We propose MoveUCB, which addresses the reachability constraint through a global target search, a persistence rule that holds the target across slots, and a transit-aware movement cost term. We prove that MoveUCB attains sublinear dynamic regret under the standard variation-budget condition of non-stationary bandits alone, with the physical parameters entering only through constants. Simulations against a dynamic-programming oracle and movement-unaware baselines show that MoveUCB attains the lowest regret with about half the travel of the closest baseline that moves.

Probabilistic question-answering systems – whether large language models (LLMs) themselves, retrieval-augmented generation (RAG), or trained multi-hop retrievers – conflate "what is known" and "how to reason" into a single probabilistic computation: hallucination cannot be eradicated, evidence chains cannot be audited, and the system answers even when it does not know. We present LatWeave, which organizes knowledge into a multidimensional knowledge lattice and compiles multi-hop QA into three deterministic operators – meet (constraint intersection), compare (lattice-order comparison), and abstain (structural abstention); LLMs appear only on the construction side (one-shot extraction) and the query-planning side, while the answer-generation path is zero-LLM, zero-task-training, and auditable end to end – so that question answering over Web-published knowledge becomes reproducible item by item. Rather than claiming across-the-board SOTA, we characterize the operating envelope of this paradigm on six public benchmarks: when knowledge is complete (MetaQA, 39,093 questions) meet chains are near-lossless over three hops (any-hit 0.9975, on par with fully supervised KBQA); on templated multi-hop home ground (2WikiMultihopQA held-out n=1,258) EM 0.865, well above published structure-augmented RAG reproductions; on open-text deep composition (MuSiQue) and extraction-coverage gaps (HotpotQA) we report degradation honestly and attribute it to causes outside the lattice-algebra layer; and when information is incomplete (IIRC) we achieve structural abstention with abstain accuracy 0.971 and leak rate 0.029. Within the operating envelope, deterministic execution pays no performance penalty, and every step on the answer path can be recomputed – precisely the source of end-to-end auditability.

Background:Flatbed scanners introduce a lateral response artifact (LRA) perpendicular to lamp motion. Because LRA varies with position, pixel value (PV), color channel, and film model, it may violate the channel-common perturbation assumption of multichannel dosimetry. Purpose:To characterize PV-domain LRA in EBT4 and EBT-XD films, compare Lewis and Full corrections, and evaluate their effects on single-, double-, and triple-channel dosimetry. Methods:Seven films were scanned at seven lateral positions using a cyclic-shift design. For each position and RGB channel, center-equivalent response was modeled as PVc = A + B * PV. Lewis correction used two films with linear interpolation of A and B; Full correction used all center-local pairs with PCHIP interpolation of PVc - PV. Performance was assessed by portrait-landscape profile consistency and dose accuracy. Results: LRA was affine in PV, asymmetric about scanner center, and dependent on film model and channel. Both corrections reduced profile differences. Mean absolute profile differences across MU levels ranged from 1.4%-5.3% for EBT4 and 2.4%-6.8% for EBT-XD. The methods performed similarly for EBT4, while Full correction reduced discrepancies for several EBT-XD double- and triple-channel methods. Residual blue-channel errors limited blue, red-blue, and triple-channel performance. At 500 MU and above, red showed the best dose agreement; at 100 MU, both corrections increased deviations for several single-channel estimates. Full correction produced smaller deviations than Lewis in several EBT-XD conditions. Conclusions:LRA should be corrected in the raw-PV domain before dose reconstruction. Full correction better preserved profile consistency and dose accuracy, particularly for EBT-XD. Because improved profile consistency did not always improve central-dose agreement, both should be evaluated separately in commissioning.

Converter terminal characteristics are central to both dynamic interactions with the network and steady-state power sharing in inverter-based microgrids. Conventional additive virtual impedance (VI) shapes these characteristics through a single virtual branch. This paper proposes virtual two-port control, which uses four coordinated transfer channels to reconstruct the terminal impedance. The connected network consequently observes the original converter through a virtually inserted twoport interface, with conventional virtual impedance recovered as a degenerate case. The proposed control transforms the original impedance through a matrix linear-fractional map. We also derive a necessary and sufficient condition for the reconstruction to be stable and proper. Among its various potential applications in microgrids, we develop three representative ones in detail: passivation with reduced control effort, uncertainty compression, and seriesshunt power-flow regulation. Simulations of VSG-controlled converters demonstrate the advantages of the proposed method over conventional VI in these applications.

Pneumonia is difficult to diagnose in older long-term care residents; multimorbidity and atypical presentations obscure signs, motivating operationally efficient objective testing. We analyzed multi-channel digital stethoscope recordings from 185 Japanese residents (73 pneumonia, 112 symptomatic without), using radiologist-confirmed chest X-rays and clinician diagnoses as supervisory signals that train convolutional neural networks, multimodal fusion, and channel-based variants with time-domain Grad-CAM interpretability. Models were evaluated with repeated patient-level cross-validation showing models with X-ray supervision outperformed clinician supervision (F1 0.729, accuracy 0.783 vs. F1 0.637, accuracy 0.711). Additionally, a three-channel selection protocol maintained performance (F1 0.736; accuracy 0.803), with two mid-thoracic sites ranking highest and Grad-CAM attention overlapping adventitious sounds. These findings indicate automated multi-channel lung-sound analysis can aid long-term care pneumonia diagnosis, with X-ray supervision being more reliable than clinical, and fewer channels preserving performance while lowering acquisition times.

This paper studies multivariate conformal prediction (CP), a distribution-free uncertainty quantification framework with finite-sample coverage guarantees. The efficiency of multivariate prediction sets hinges critically on the residual geometry encoded by the nonconformity score, while existing minimum-volume methods rely on quantile thresholds that ignore tail residual severity and implicitly bind geometry learning to coverage level. We propose a tail-aware geometry learning framework for conformal ellipsoids that decouples tail sensitivity in geometry learning from the final coverage guarantee. Using a two-split design, we learn the metric matrix via volume minimization under a CVaR constraint on an estimation split, then apply standard conformal calibration on a held-out calibration split. The resulting problem is convex and admits a bounded-reweighting interpretation that prioritizes high-residual samples. Moreover, we theoretically characterize the trade-off between ellipsoidal volume and tail severity. Experimental results demonstrate the effectiveness of the proposed method.

Diffusion language models generate text through iterative denoising, exposing intermediate trajectories before final answers are produced. We identify a recurring reasoning failure, stable-but-wrong lock-in, where an answer stabilizes early around an incorrect value while substantial denoising remains. Surface-level decoding signals such as confidence, entropy, margin, and answer stability are insufficient to reliably distinguish correct from erroneous lock-in. We formulate selective reasoning repair as a lightweight test-time planning problem and propose LOCKR, a hidden-state trajectory-guided planner that decides when to allocate additional computation, expands a structured set of targeted repair branches, and selects the most promising continuation using trajectory-aware verification. Across two diffusion language models and three mathematical reasoning benchmarks, hidden-state trajectories consistently outperform surface signals and single hidden snapshots for both wrong-lock-in detection and repair selection. On natural evaluation distributions, LOCKR yields absolute accuracy gains of 2.21–5.37 percentage points across all five evaluated settings, with repair rates ranging from 22% to 41%. These results establish hidden diffusion trajectories as actionable signals for selective test-time reasoning repair.

Aerial radiation surveys achieve higher sensitivity when the radiation detector is held close to the ground. Detector sensitivity falls off roughly with the inverse square of the distance to the source, so a detector flown high is slower to reach a given minimum detectable activity. Flying the vehicle low puts the propellers near the ground, where downwash can disturb the surveyed area and resuspend contaminated particulates. Tether suspension decouples the detector from the vehicle altitude, but leaves the payload unactuated and only indirectly controllable. We therefore present a vision-based control approach for an aerial sensing payload suspended on a tether beneath a heavy-lift drone. Because a survey plan is decided as radiation detections arrive, we design a pilot aid for commanding the survey trajectory manually with a handheld transmitter. The controller regulates the payload, rather than the vehicle, onto that trajectory. The system uses onboard sensors with a downward-facing camera fixed to the drone body tracking a ring marker on the payload. A four-state Kalman filter estimates the tether swing angles and rates from payload bearing measurements, and a linear quadratic regulator with integral action takes the payload position as the regulated output. In outdoor flight tests under wind, the payload-aware controller reduced payload tracking error during transit by 20% when compared against a vehicle-referenced baseline, with the cost of higher peak error on arrival at a waypoint.

Embodied navigation requires spatial representations that generalize across unseen environments, yet collecting large amounts of annotated data from real 3D environments is difficult. We propose NaviScale for semantic-map-based object navigation (ObjectNav), whose predictor can be trained on pairs of partial and complete semantic maps without reconstructing a complete 3D environment for every training sample. The framework generates large-scale semantic map training data by composing floorplans of real homes with room-level semantic and obstacle maps extracted from MP3D and HM3DSem. NaviScale increases data diversity in two ways: inter-room scaling increases floorplan-level structural diversity, while intra-room scaling fills each fixed floorplan with different combinations of room maps matched by room category. Visibility through Ray Casting (VisRC) converts the composed maps into partial observations that account for field of view, sensing range, and occlusion. The resulting dataset contains 192,000 semantic maps generated from 24,000 floorplans associated with 12,794 properties. With 300k training iterations and the training and inference settings described in this paper, the system reaches 64.3% SR and 34.8% SPL on HM3D, together with 43.1% SR and 16.8% SPL on MP3D, without changing the prediction architecture. Additional experiments evaluate the quality of the composed maps, the effects of semantic-segmentation errors, and deployment on a physical robot.

We address adaptive computation in 3D medical image segmentation: instead of designing another backbone, we ask how much spectral mixing each network stage needs and let optimization answer. We derive FHEAT, a two-parameter operator family, from the discrete cosine transform (DCT) solution of a fractional heat equation. A fractional order alpha and a diffusion strength D govern the operator, and at D=0 it is exactly the identity. Reparametrized by the semigroup time tau = D*alpha, same-resolution instances compose exactly, so any distribution of diffusion across same-resolution stages amounts to a single Sobolev-type regularizer of learned strength. This identity limit lets the optimizer of each layer, not the designer, decide whether global mixing is needed and how sharp it should be. We instantiate FHEAT in a lightweight U-shaped architecture (Light-UNETR) paired with a Kolmogorov-Arnold mixer (KAN3D) with adaptive rational activations, yielding FHEAT-Seg. At 5% to 20% label rates on three public benchmarks, training produces gradient-driven spectral sparsification: seven of the eight stage-level operators drive D to zero, and the survivor saturates at the sharpest low-pass (alpha ~ 0.9) in the decoder layer feeding the semi-supervised attention map. The retired layers become exact identity shortcuts at inference, cutting FLOPs from 4.29G to 0.90G (a 79% drop) at 0.975M parameters. Under a standard semi-supervised protocol, FHEAT-Seg reaches Dice scores of 90.47% (left atrium), 78.79% (Pancreas-CT), and 81.90% (BraTS 2019), ahead of five semi-supervised methods and the Light-UNETR baseline. The large variant also surpasses Light-UNETR-L under full supervision (Dice 93.09%, 85.11%, and 87.19%) with 2.851M parameters and 55.75G FLOPs. These results suggest that the allocation of spectral computation is a learnable property of optimization dynamics, not a manual design commitment.

Topology optimization (TO) remains computationally intensive due to repeated finite element analysis (FEA) evaluations required at each iteration. While neural network-based surrogates offer potential acceleration, existing approaches often suffer from gradient inconsistency between predicted objectives and sensitivities, leading to optimization instability. This work presents KATOsuper, an objective-agnostic framework that couples neural-reparameterized topology optimization with a Sensitivity-Consistent Fourier Neural Operator (SC-FNO). The framework employs the forward_split architecture, which derives deployed sensitivities via automatic differentiation through the predicted objective field and thereby preserves consistency between the predicted objective and the gradient used for optimization. The case studies include three 2D benchmark problems and three 3D structures considering compliance or stress minimization. A physics-informed multi-channel input encoding with Fourier position embedding enables resolution-invariant learning, supporting zero-shot extrapolation beyond the training resolution, with useful performance at moderate scaling factors and topology-preserving exploration at up to 64x without retraining. The framework extends to 3D through KATO3D, featuring novel KANConv3D blocks with learnable B-spline activations. KATOsuper demonstrates 15–110x deployment-time speedup over MATLAB baselines while maintaining competitive optimality, with the clearest gains observed in complex 3D and stress-optimization cases. The insight that sensitivity direction matters more than magnitude enables robust optimization even with approximate physics evaluation, extensible to other differentiable physics-driven design objectives.

Kerr photonic integrated circuit optical parametric oscillators (PIC OPOs) provide a promising platform for coherent visible-light generation, offering broad wavelength accessibility and low frequency noise for applications in quantum information processing, precision metrology, spectroscopy, and sensing. However, achieving large fiber-coupled output powers and high conversion efficiency remains challenging, impacting the practical deployment of these sources. Here we demonstrate optical injection locking of a commercially available visible Fabry-Pérot (FP) laser diode to a PIC OPO, realizing a hybrid laser that combines the complementary strengths of nonlinear frequency conversion with those of semiconductor laser technologies. Using 25 $\mu$W of injected PIC OPO signal, we generate coherent visible-light at 635 nm with a fiber-coupled output power of $\approx$ 40 mW and side-mode suppression ratio $\gtrsim$ 36.3 dB. The injection-locked FP laser faithfully inherits the low frequency noise characteristics of the PIC OPO signal, achieving a frequency noise floor of 714 Hz$^2$/Hz. We further demonstrate $\gtrsim1$ GHz wide locking bandwidths that enable stable injection locking without active PIC OPO stabilization, together with coarse wavelength tunability over 6 nm while preserving both high output power and spectral purity. Our results establish optical injection locking with nonlinear light sources as a compelling option for coherent, high-power, and wavelength-agile integrated coherent visible sources.

Compiler optimizations traditionally rely on handcrafted heuristics that often fail to generalize across programs and architectures. We investigate whether large language models can participate in compiler optimization through a verification-centered systems architecture that couples generative rewriting with formal equivalence checking. Using lazification in LLVM IR as a case study, we fine-tune a code-centric LLM on transformations produced by Wyvern and embed Alive2 into a feedback loop that enforces semantic preservation for every generated rewrite. Correctness is enforced externally as a runtime control layer rather than learned implicitly. During inference, candidate transformations are symbolically validated and regenerated when necessary, ensuring accepted rewrites satisfy formal constraints. On the LLVM test suite, the fine-tuned model reproduces core optimization behaviors while applying fewer transformations overall. Although Wyvern remains faster on most benchmarks, 9.8% achieve comparable or improved runtime under the learned system, with no semantic violations observed. Verification overhead remains bounded and convergence stable. These results demonstrate that generative AI components can be safely integrated into compiler pipelines through deterministic validation and structured feedback, offering a scalable architectural pattern for trustworthy AI-driven software infrastructure.

Open-ended queries in modern Retrieval-Augmented Generation (RAG) are increasingly "diffuse," requiring a large set of documents to be assembled into a finite LLM context window. To ensure retrieval quality, systems use fast dual-encoders and more expensive cross-encoders (CEs) to score candidates. However, the CE budget $B$ is strictly bounded by latency and is often smaller than the context window capacity $k$. This mismatch makes standard reranking structurally flawed: it wastes compute verifying obvious top candidates while ignoring relevant documents further down the initial ranking. To address this, we introduce BoundaryMORPH, a novel algorithm that allocates CE budget specifically for the LLM's context capacity $k$. Using a Gaussian Process, BoundaryMORPH treats the initial dual-encoder ranking as a structural prior and intelligently spends CE calls on resolving top-$k$ set membership at the boundary, rather than seeking a single most-relevant document. Information from each CE call propagates to unscored documents, maximizing the utility of the budget. We demonstrate that BoundaryMORPH achieves state-of-the-art set retrieval quality across multiple models and datasets with open-ended queries ($+5.4$ nCG@100 over the strongest baseline).

Electric-field control of magnetic order and valley polarization holds great promise for spintronic and valleytronic applications. However, achieving such electrical modulation remains a fundamental challenge in two-dimensional (2D) van der Waals (vdW) magnets. Herein, via first-principles calculations, we verify that electrically tunable interlayer magnetic order and valley polarization can be realized in Janus VSSe bilayers with different interfaces. Although the magnitudes of dipole moments within each constituent Janus monolayer are identical, the vertical built-in electrostatic potential difference $\Delta\phi$ across the bilayer depends strongly on the orientations of these dipoles, which originates from the distinct interfacial configurations. As a consequence, the VSSe bilayers with different interfaces possess distinct interlayer magnetic couplings and show dramatically varied responses to external electric fields. For Se-S interface, reversible electric-field switching between antiferromagnetic (AFM) and ferromagnetic (FM) states can be achieved due to the competition between itinerant-electron-mediated FM exchange coupling and interlayer $\Delta\phi$-dependent AFM/FM super-superexchange interactions mediated by interfacial Se and S atoms. For Se-Se interface, both the valley polarization and spin splitting can be effectively reversed by the out-of-plane electric field, realizing an all-electric-field controlled anomalous valley Hall effect. Our results demonstrate that the interface of the magnetic Janus bilayer could act as an additional degree of freedom to tune the electronic, magnetic, and valley properties in 2D vdW materials. The Janus VSSe bilayers are identified as a promising platform for the design of low-power spintronic and valleytronic devices.

In this paper, we consider the strong operator topology (SOT) on the space $\mathcal{L} (X)$ of continuous linear operators on an infinite-dimensional Fréchet space $X$. The existence of SOT-dense subspaces as well as of SOT-closed infinite-dimensional subspaces inside the family of all non-cyclic operators is established. Other classes of operators, such as non-dense range or non-injective operators, are also studied in this regard. Moreover, we prove for the sequence space $\ell_p$ $(0 < p < \infty)$ that the family of all non-cyclic members of $\mathcal{L}(\ell_p)$ contains an isometric copy of $\ell_p$.

In this paper, we prove that $\mathbb{Z}$ is first-order definable in the ring of integers $\mathbb{Z}^{\text{ab}}$ of the maximal abelian extension $\mathbb{Q}^{\text{ab}}$ of $\mathbb{Q}$, which implies that the first-order theory of $\mathbb{Z}^{\text{ab}}$ is undecidable. More generally, writing $i = \sqrt{-1}$ and $\mathbb{Q}^{\text{tr}}$ for the field of all totally real numbers, we prove new definability and undecidability results for rings of integers of subfields of $\mathbb{Q}^{\text{tr}}(i)$, focusing especially on fields which contain infinitely many roots of unity. The key ingredient for these results is that there is a parameter-free positive-existential formula which defines the roots of unity $\mu(\mathcal{O}_L)$ inside $\mathcal{O}_L$ for every field $L\subseteq \mathbb{Q}^{\text{tr}}(i)$.

Subgraph sampling reduces the training cost of large-scale graph neural networks, but sampling criteria may overlook the geometric roles of edges. We propose a resistance-curvature-guided sampling framework built on ERC-LG, a curvature approximation method for large-scale graphs. ERC-LG combines Johnson-Lindenstrauss projections with regularized multi-GPU batched conjugate gradient solvers, avoiding explicit Laplacian pseudoinverse computation and full embedding storage. The resulting curvature informs node- and edge-sampling probabilities for constructing GNN training subgraphs. Experiments show numerical agreement with pseudoinverse-based curvature and reduced runtime compared with CG-only computation. ERC-LG-based sampling variants achieve the highest mean accuracy on six of seven real-world datasets in downstream node classification.

Spatial transcriptomics (ST) measures gene expression in tissue context, but dense capture grids can be costly and may repeatedly sample morphologically similar regions. Most active learning strategies were developed for categorical labels and independent samples. We conduct a retrospective pool-based benchmark of active learning versus uniform Random sampling for ST, where expression vectors are high-dimensional and continuous and candidates are spatially correlated. Using two fully profiled public ST cohorts, we mask candidate expression vectors and simulate multi-round selection with uncertainty-based Monte Carlo dropout (MC-dropout) and temporal output discrepancy (TOD), and diversity-based CoreSet and TypiClust-inspired selection. We compare 160 completed configurations at 5%, 10%, 30%, and 50% of the fold-wide training spot pool under patient-level cross-validation, with a separate full-label reference. Within each budget, strategies share the selection schedule, morphology-to-expression predictor, and optimization protocol. We assess mean per-gene within-slide Pearson correlation coefficient (PCC), expression-cluster agreement, and Moran's I fidelity. On HER2-positive breast cancer, pooled mean PCC differences from Random across the four active strategies were -0.0176, -0.0117, +0.0056, and +0.0057 at 5%, 10%, 30%, and 50%, respectively. On cutaneous squamous cell carcinoma (cSCC), three strategies were below Random at 5%, and all four were below Random at 10%. On HER2-positive breast cancer, CoreSet and MC-dropout had lower PCC but higher expression-cluster agreement than Random at the two smallest budgets; this pattern did not reproduce on cSCC. Under the reported fixed training horizons, the evaluated active strategies do not consistently improve on Random at small budgets, and rankings depend on the evaluation measure.

DNA sequencing of the gut microbial community shows promise for cancer detection, but questions remain about the generalizability of results across studies. We propose BreCol, a benchmark of 2,040 16S rRNA gene sequencing runs across 26 studies spanning breast cancer, colorectal cancer, and healthy cohorts. Train-test splits are made within pre-2023 studies, while holdout evaluation uses studies from 2023 onward, reflecting temporal separation from training data. Classical models reach test/holdout AUCs of 0.77/0.60 for cancer diagnosis and 1.00/0.83 for cancer type prediction. We train the models on both cancer types simultaneously and find that colorectal cancer is often easier to detect than breast cancer. We also evaluate two deep learning models: HyenaDNA, a long-range sequence model that pools hidden states for classification, and SetBERT, a transformer that produces contextualized embeddings over sets of reads. Both deep learning models underperform the best classical methods on holdout data, though tuning training set size and the classification head yields modest gains. Our classical pipeline uses unsupervised clustering to derive features from tetramer frequencies, preserving within-run compositional signal and achieving near state-of-the-art performance without relying on taxonomic assignments. BreCol data and associated code are publicly available.