Papers

We experimentally explore the collective behavior of the nematode T. Aceti inside flexible boundaries showing the emergence of previously unreported states. These nematodes have been previously shown to be able to synchronize their body oscillations in a favorable condition of confined space. In this collective state, they are able to exert a strong pushing force which we exploit to study how their collective motion could deform a pliable boundary. Using a novel experimental technique, we were able to simulate a non-rigid border at a liquid-liquid-gas triple point. We discovered a state with periodic boundary deformations which are due to the synchronized oscillations of the nematodes. More interestingly, we found a novel state where the nematodes are able to form multiple protrusions in the border which oscillate, rotate, merge, and split. We propose a numerical model that is able to reproduce some of the observed parameters of this state as well as offer an analytical insight into protrusions dynamics. Our research opens a pathway for exploring the interaction between active matter and flexible interfaces with a new kinds of active matter agents forming dynamic non-chaotic border deformations.

Wildlife-vehicle collisions remain a significant road safety concern, particularly during nighttime and low-visibility conditions when RGB-based perception systems are often unreliable. Thermal imaging offers a promising alternative for detecting animals under poor illumination. However, the limited availability of annotated infrared animal datasets restricts the development of robust deep learning-based detection models. This paper investigates synthetic thermal image generation as a scalable approach for real-time animal detection under low-visibility conditions. A subset of 514 annotated visible-spectrum animal images from the NTLNP dataset is translated into synthetic thermal representations using CycleGAN-Turbo, while a limited real thermal dataset of 60 images is expanded through thermal-focused augmentation. Multiple object detection architectures, including YOLOv8, YOLOv9, YOLOv10, and RT-DETR, are trained independently on synthetic and real thermal datasets and evaluated using precision, recall, mAP@0.5, mAP@0.5:0.95, model size, and inference latency. Experimental results show that synthetic thermal images provide competitive detection performance, with RT-DETR achieving the highest synthetic-data mAP@0.5 of 0.9613. Models trained on augmented real thermal data achieve the strongest overall performance, with YOLOv10s obtaining 0.9879 mAP@0.5 and 0.9571 mAP@0.5:0.95. Computational analysis further indicates that lightweight YOLO variants provide favorable inference latency, supporting their potential for real-time deployment. These findings demonstrate that synthetic thermal imagery can reduce dependence on scarce infrared datasets and support the development of efficient animal detection systems for future vehicle-mounted wildlife collision mitigation applications.

Engineered dissipation corrects errors, yet a memory's phase decays. What the memory's gap measures is not settled. Here we show that when each jump resets one basis state, the wiring can set the gap time while the phase keeps its rate. Under dephasing the gap time spans two orders across wirings; the phase does not. In this class, a contrast on two or more qubits fades faster than a bare qubit at any distance; no such pump power substitutes. The ceiling is two over the qubits the contrast spans. A two-branch pump can escape it at distance three; a two-clock test runs on published data.

A two-hop question makes a language model retrieve twice: once to produce a bridge entity, and once to retrieve with it. Transformers resolve that entity in their early layers. Hybrid models replace most of the attention with a recurrent state, so where the key becomes usable, and where it is spent on an answer, is not known. Answering "Where is the ball?" from "the ball belongs to Alice" and "Alice is in the garden" turns on Alice, a name the question does not mention. A model could reach garden through Alice, the key it computed, or through where the fact sits in the prompt. Across twelve models, dense and hybrid, we move a hidden state from one story into another where the two routes lead to different places, and read off which place the model gives. An attention layer converts the key in every model we tested, and crossing that layer removes its usable effect, all of it where no attention follows. In sequential hybrids this makes the answer a handoff: recurrent layers carry the key forward, and attention spends it. Retrieval does not always stop there. Writing a different fact into the memory of the recurrent layers that follow the last attention layer can move the answer toward that fact, multiplying its odds by 1.3 to 2.7, and which hybrids do this is not settled by their architecture or training. That read is addressed by the key, not by position: recurrent state in a hybrid is not only a carrier, but a memory that later layers can query.

Inferring quantities of interest (QoI) from data is a central task in Science and Engineering. In such contexts, we often have access to both quantitative data and qualitative data. Quantitative data may be represented by noisy sensor measurements, simulation data, re-analysis data; qualitative data may be in the form of text descriptions of experimental setups, expected experiment outcomes, and human-perceived system behaviours. The task we address in this paper is the following. Given a training set of paired qualitative text and quantitative QoI data, we learn to exploit the inherent correlation between the two modalities to learn a highly informative data-driven natural-language-conditional Bayesian prior, such that when presented with a new physical system, we can coherently combine (i) the training dataset, (ii) qualitative text describing the new system, and (iii) a small number of noisy sensor readings from that new system, to perform inference and uncertainty quantification (UQ) over the QoI. To achieve this task, we develop two parallel approaches, one uses conditional diffusion and the other conditional autoencoders, and compare both against classical Bayesian methodology, unconditional generative models and deterministic supervised methods. Each approach has specific strengths and tradeoffs; conditional autoencoder offers theoretical tractability, allows for fast posterior sampling, and provides better-calibrated UQ, whereas conditional diffusion is explored for greater expressiveness and capturing complex posteriors with irregular QoI fields. The approach is tested on the steady-state heat equation, damped Helmholtz equation, and UK weather reanalysis data.

Graph convolutional networks propagate information by repeated local aggregation through a graph shift operator; i.e., a $K$-layer network reaches $K$ hops neighborhood. On the one hand, such spreading can lead to oversmoothing. On the other hand, long-range dependencies demand the depth. Transporting information on long distances and without attenuation requires the shift to distinguish a direction of flow, which a symmetric operator cannot perform but a directed one can fulfill. A natural way to extract pure-directionality is to take the skew-symmetric part of the shift operator through the Cartesian split, which, however, generally does not commute with the shift itself, meaning that the filters built on it are not shift-invariant. We instead use the spectral conjugate; i.e., the image of the operator under $\tau:z\mapsto \bar{z}$, which commutes with the shift and splits it into a dissipative and a non-dissipative part. Two filter families follow: a sum filter, whose non-dissipative component transports signal without energy loss, and a ratio filter, ratio in the pair of components rather than polynomial in the shift. Both arise from non-holomorphic kernels, placing them outside the holomorphic class underlying classical spectral convolution. On the directed cycle, the ratio filter becomes an IIR filter with global impulse response, for which we prove a long-range reach gap against every degree-$K$ polynomial filter. Chebyshev reparameterization gives stable vertex-domain layers with real coefficients, yielding TwinS-GCN, which solves graph transfer tasks at reduced depth and is competitive with state-of-the-art graph convolutional networks on node classification benchmarks.

Multi-agent systems built on large language models (LLMs) are largely homogeneous, as their agents behave alike even across distinct LLMs. We show that when such agents act concurrently without communication, they collide on targets they must split and diverge on targets they must take together, a double failure we term the symmetry trap. Theory of Mind (ToM), widely used for coordination without communication, cannot escape this trap, since homogeneous agents form the same prediction of one another and respond to it in the same way. We propose Theory of Scene (ToS), a training-free reasoning schema in which each agent reads its public role, the only difference between the agents, and the task context they all observe. Homogeneous agents thereby derive one division of labor, each taking the part its role fixes, which turns homogeneity from the cause of the trap into the cure. ToS reads the role together with the scene through role gating, which determines whether ownership overlaps or is already divided, and the task context through task coupling, which infers whether the team must converge on each target, divide it, or take its stages in turn. We evaluate on DivvyBench, a controlled environment we introduce, whose target types make an episode Competitive, Cooperative, or Mixed across Tabletop, Airspace, and Household scenarios, and on two established agentic benchmarks, GovSim and Overcooked. ToS outperforms all six baselines on every benchmark, and each baseline falls far behind it in at least one setting. Against ToM given the same inputs, ToS raises the DivvyBench success rate from 71.1% to 99.6%, the GovSim total gain from 207 to 400, and the Overcooked level-normalized throughput from 1.41 to 1.67.

The effect of a porous coating on the water-exit dynamics of a circular cylinder is investigated experimentally. Cylinders comprising a $30~\mathrm{mm}$ solid core surrounded by a $10~\mathrm{mm}$-thick porous foam layer with pore densities of 10, 20, and 60 pores per inch (PPI) are compared with a smooth cylinder of the same outer diameter. Experiments are performed over $1.50\times10^{4}\leq Re\leq5.00\times10^{4}$ and $0.37\leq Fr\leq4.08$. Time-resolved particle image velocimetry characterizes the wake and reconstructed pressure field, while proper orthogonal decomposition (POD) examines its energetic organization. All porous cylinders reduce free-surface elevation relative to the smooth cylinder, with the largest reduction generally obtained for 60 PPI. The submerged wake, however, is not simply weakened. Porous cases generally develop larger circulation and vortex area, with a non-monotonic dependence on pore density and operating condition. The 10 PPI cylinder frequently produces the largest vortex region, whereas 60 PPI generates stronger negative pressure within the vortex and the largest integrated vortex-associated pressure deficit for $Fr=1.02$, $2.00$, and $4.08$. During the submerged approach to contact, this pressure deficit co-evolves strongly with circulation, with Pearson correlation coefficients typically between $0.90$ and $0.98$. POD shows that 9–18 modes recover 95% of the modal energy, with 20 PPI requiring only 9–10 modes. Overall, porous coating reorganizes wake strength, extent, pressure signature, and energetic structure while reducing the free-surface response.

We prove that there is an absolute positive constant $c$ such that every bipartite graph with $N$ vertices in each part and at least $N^2/2$ edges contains a complete bipartite graph $K_{t,t}$ whenever $N\ge c\, 2^{t}$. This improves the classical Kővári-Sós-Turán bound requiring $N$ of order $t\,2^t $. As a consequence, the diagonal bipartite Ramsey number has upper bound $b(t,t)=O(2^t)$, improving the previous best bound $b(t,t) = O(2^t\, \log t )$ due to Conlon. The proof was found by GPT-6 Astra, and the method will probably have further applications.

We give a strict generalization of the LLL algorithm over number fields, based on the reduction theory of $\GL(n)$ over the adele ring of a number field. Our algorithm is free of heuristics, with rigorous bounds on output quality and complexity. As a consequence, we obtain a hierarchy of reductions from module-(H)SVP to ideal-HSVP, an example of which has runtime and approximation factors subexponential in the field degree. More importantly, we uncover a close connection between structured lattice reduction and a Diophantine approximation over number fields.

Multi-human multi-robot (MH-MR) teams combine robotic autonomy with human expertise, but effective task allocation requires coordinating scarce, dynamically available human support with distributed robot execution. Limited robot-robot and human-robot communication further complicates this coupling by delaying information exchange and supervisory intervention. We introduce CommHG, a communication-aware heterogeneous graph learning framework for decentralized MH-MR task allocation. CommHG represents coupled human-robot-task interactions through local graphs conditioned on information availability and age. Learned communication actions enable robots to decide when and which operator to query, and when to share information with peers. Allocation and communication are jointly optimized through cooperative multi-agent reinforcement learning, allowing the team to acquire useful information while managing limited communication and supervisory resources. We also introduce a benchmark integrating heterogeneous humans and robots, dynamic tasks and operational states, and constrained robot-robot and bidirectional human-robot links. Experiments across heterogeneous teams with up to 16 robots, 6 humans, and 112 tasks show that CommHG improves timely weighted mission completion as coordination scale increases, outperforming the strongest baseline in the Large scenario.

Neural network training is inherently stochastic, with factors such as weight initialization leading to distinct models despite comparable predictive performance. This phenomenon is commonly associated with the Rashomon effect, which describes the existence of multiple near-optimal models for the same task. Although the Rashomon effect has received increasing attention, it remains unclear whether different sources of training stochasticity contribute similarly or differently to its manifestations. In this work, we present an empirical study of the Rashomon phenomenon along three complementary dimensions: solution-space multiplicity, predictive multiplicity, and decision-basis multiplicity. These dimensions are quantified through the size of the empirical Rashomon set, predictive ambiguity, and agreement between XAI attribution maps, respectively. By independently controlling three standard sources of stochasticity, namely weight initialization, mini-batch data ordering, and dropout, we isolate their respective contributions to each dimension of the Rashomon phenomenon. Experiments on tabular and image classification benchmarks reveal that these sources affect the three dimensions in different ways. In particular, larger empirical Rashomon sets do not necessarily correspond to greater predictive disagreement or lower explanation agreement, indicating that solution-space, predictive, and decision-basis multiplicity capture complementary rather than interchangeable aspects of the Rashomon effect. Overall, our results show that training stochasticity influences not only predictive performance but also the stability of predictions and explanations, highlighting the importance of identifying the specific sources of stochasticity responsible for different manifestations of the Rashomon phenomenon when assessing the reliability, reproducibility, and interpretability of neural network models.

In safety-critical applications, deployment uses a single policy, whose performance and constraint satisfaction should hold directly rather than only for an average or mixture of training policies. This motivates last-iterate guarantees in constrained reinforcement learning. Recent progress has established such guarantees in exact-gradient or tabular online settings, yet scalable results for structured large-state problems remain open. We develop a general, statistically efficient framework for last-iterate convergence in structured Constrained MDPs (CMDPs). Our analysis separates contraction of the regularised primal-dual dynamics from actor approximation and statistical errors in policy evaluation under online exploration. This enables model-free on- and off-policy learning with structured function approximation: optimistic policy evaluation avoids explicit transition-model construction, while a compact parametric actor avoids maintaining mixtures or histories of past policies. We instantiate the framework for linear CMDPs and general function approximation, obtaining representation-dependent complexity and improved target-accuracy dependence over prior optimistic regularised primal-dual analyses. We further validate the stabilising effect predicted by our theory on a synthetic linear CMDP: the regularised method exhibits stable last-iterate behaviour, whereas its unregularised counterpart shows larger oscillations.

Finite computational resources force a tradeoff between automatic System 1 processes and costly System 2 thinking. Large language models (LLMs) can spend extra computation on hard problems, yet direct answers struggle even with counting, an elementary operation humans and animals perform automatically. We ask why this requires thinking in LLMs. Evidence integration has long been used in psychology and neuroscience to probe decision-making. Our evidence-integration task presents one letter per conversational turn and asks which of two target letters appeared more often. A running count difference solves the task optimally by weighting every letter equally; tokens at each turn could represent and update this difference. Direct responses instead weighted evidence unevenly, with strong recency effects, and assigned less probability to the correct answer as difficulty increased. Thinking improved performance and made integration weights nearly uniform, yet final-query attention remained concentrated on the sequence ends in both modes. Reasoning trajectories showed models revisiting input, recounting letters, and checking intermediate counts that informed the answer, suggesting that thinking constructs the accumulated count that direct responses lack rather than reading out one already formed. Reasoning-token costs grew with the number of letters far more than with coherence. Outcome feedback did not bring this computation into direct responses: under in-context reinforcement learning (ICRL), performance deteriorated over repeated games and recency effects strengthened, yet models grew more confident. Humans and animals amortize such computations into automatic processes, whereas current LLMs still pay for them with thinking on every trial. Which operations learning can make directly available remains central to how future models allocate computation.

Lebesgue's universal covering problem asks for the minimum area convex planar region capable of containing a congruent copy of every planar set of diameter at most one. In this paper, we develop an exact dual formulation of this problem through a minimum diameter analogue of Bellman's lost-in-a-forest problem. For a compact convex forest $F$, we define the critical escape diameter $D(F)$ as the minimum diameter of a closed path whose trace cannot be placed, under any rigid motion, entirely in the interior of $F$. We prove attainment of this minimum and establish a diameter cover/escape duality showing that the normalized body $D(F)^{-1}F$ is a Lebesgue universal cover. Consequently, the Lebesgue universal covering constant admits the exact representation \[ \mathcal L=\inf_F\frac{\operatorname{Area}(F)}{D(F)^2}, \] where the infimum ranges over compact convex planar bodies with nonempty interior. By reversing the rigid motion, we further characterize escape as intersection of a fixed path with every oppositely transformed boundary of $F$, and derive equivalent continuous curve, convex body, and support function optimization formulations. To make the infinite dimensional problem computationally tractable while retaining rigorous control of approximation error, we discretize the compact configuration space by an $\eta_m$-net and formulate the resulting problem as a minimum diameter traveling salesman problem with neighborhoods; for polygonal forests, an exact mixed-integer second-order cone formulation is obtained. We prove the quantitative certification, which yields convergent, rigorously certified universal cover bounds. The framework replaces finite tests of prescribed constant width shapes by a unified optimization over the full configuration space and extends naturally to other congruent and translative universal cover problems.

Brain-computer interfaces (BCIs) have emerged as transformative technologies that enable direct communication between the brain and external devices. Among various BCI paradigms, EEG-based motor imagery (MI) has gained prominence due to its simplicity, non-invasiveness, and potential to restore motor function and facilitate rehabilitation for patients with motor impairments. This paper presents a comprehensive review of the most practical processing algorithms developed over the past decade for decoding brain sensorimotor cortex signals. Specifically, this paper discusses the integration of artificial intelligence (AI)-based algorithms, particularly machine learning and deep learning techniques, and their contributions to improving the performance and efficiency of MI-BCI systems in detail. Furthermore, the paper reviews state-of-the-art hardware platforms and emerging converging technologies, including system-on-chip (SoC) architectures, application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), wearable devices, the Internet of Things (IoT), and augmented/virtual reality (AR/VR), and discusses their integration with advanced signal processing algorithms to enable next-generation MI-BCI systems. By highlighting current achievements of EEG-based MI-BCI technology and predicting future research directions that could further enhance real-time capabilities, this paper aims to provide valuable insights for researchers and practitioners, fostering innovation in high-performance EEG-based MI-BCI systems.

Graph Neural Networks (GNNs) are widely used for representation learning on graphs, but most methods assume static topologies, making them inefficient on evolving networks where edges change over time. Existing dynamic approaches either model graph evolution through temporal GNN architectures without focusing on efficient dynamic maintenance, or are restricted to linear propagation models based on Personalized PageRank. In this work, we study how to efficiently maintain node representations for non-linear GNN propagation under edge insertions and deletions. The propagation has no learned parameters, and only a classifier applied afterward is trained. For a broad class of standard activation functions, we develop a residual-based dynamic algorithm that selectively propagates local errors via push operations, maintaining an approximation to the evolving fixed point without full recomputation. We prove that our method achieves amortized $O(1/\epsilon)$ update time per graph change under a degree-normalized error guarantee. Our approach uses a potential-based analysis in a degree-scaled norm and, in contrast to prior work on the linear case, requires no randomness assumptions on either the update sequence or the input vector. For the linear special case, we additionally provide an exact dynamic algorithm via low-rank matrix inverse updates. Experiments on benchmark datasets show that incorporating non-linearity improves accuracy while preserving efficient update performance, yielding a scalable and theoretically grounded method for maintaining this propagation on dynamic graphs.

Millimeter-wave (mmWave) radar and RGB-D cameras can record cardiac and respiratory waveforms continuously and without contact, but labeled recordings remain scarce because every label requires a supervised acquisition session. Self-supervised pretraining can exploit the unlabeled signals, yet contrastive methods depend on signal transformations and negative pairs whose validity is uncertain for cardiorespiratory data, where time warping changes breathing rate and distant windows can share the same physiological state. We present Phenomenon-Graph JEPA, a joint-embedding predictive architecture that learns from four processed one-dimensional streams without negative pairs or synthetic augmentation in its base configuration. Each stream is encoded by a temporal convolutional branch and a band-limited spectral branch. During pretraining, the model predicts stopped target embeddings along typed edges, which connect streams assigned to the same physiological phenomenon, and forward in time within a state episode. We treat this physiological typing as a testable hypothesis and compare it with wrong-edge and all-pairs prediction graphs. In the OMuSense-23 dataset, pretraining improves label-efficiency area over matched supervised training by 3.91 percentage points (95% interval 2.08 to 5.80, Holm-adjusted p = 0.006), and by 3.74 points under a second configuration evaluated on the same test participants. However, the wrong-edge and all-pairs controls do not establish a benefit from physiological typing. Optional Takens-inspired delay coordinates improve a validation comparison with learned history, whereas two wrist-only WESAD protocols do not establish a pretraining advantage. The study therefore separates the measured benefit of predictive representations from the physiological prior used to organize their training.

Transformer-based language models perform well on symbolic tasks, yet it remains unclear whether they learn generalizable rules or rely on statistical shortcuts. Mechanistic studies link algorithmic behavior to structured internal representations, motivating the hypothesis that robust reasoning benefits from separating values from the types that control their manipulation. Can making this separation an architectural primitive improve the learnability and generalization of logical mechanisms? We introduce STRAT (STratified Registers And Types), which partitions the residual stream into orthogonal Data and Type subspaces and uses Type-based attention and gating to govern Data transformations. Controlled arithmetic ablations identify three failure modes associated with data-control interference: the Linear Trap, Gradient Wall, and Open Gate Trap. Mechanistic analysis reveals interpretable logical structure, and in arithmetic, STRAT reduces median OOD error 35-fold relative to a Transformer baseline. On each of 11 datasets spanning 10 tasks, STRAT outperforms the Transformer baseline in mean accuracy, by 26 percentage points on average, with both models trained from 10 base examples per dataset using identical task-specific augmentation where applicable. Under distribution shift, STRAT's mean accuracy drops by only 2.39 percentage points, compared with 11.75 for the Transformer.

Learning the structure of latent factor models involves two central challenges: (1) estimating the number of latent factors and (2) learning the support of the mapping from latent variables to observed variables. This is especially challenging for nonparametric regimes and nonlinear settings. We propose a method for latent structure learning, based on pairwise dependence measures on the observed variables using a graph-theoretic representation. We show that both the number of latent factors and nonlinear mapping structure can be identified from the distribution of observed variables under mild structural assumptions. Unlike prior work restricted to linear correlations, we establish identifiability and consistency for a general class of dependence measures under nonlinear factor models. This motivates a Dependence Thresholding (DT) algorithm, which jointly estimates the number of latent factors and nonlinear mapping structure from observational data alone. We pair this with a neural network architecture constrained by the nonlinear mapping structure, to recover the nonlinear function. Through simulation studies, we show that the DT algorithm is accurate in practice, even when using flexible methods such as neural networks, and exhibits robustness against violations of its assumptions in high-dimensional settings.

A small share of online creators earn fortunes while most earn almost nothing. What separates them: skill, luck, or copying what works? This paper measures what YouTube vloggers actually produce: 3,728 thumbnails, 385 hours of speech (2.9 million words), and video titles and tags, linked to complete view histories of 56,419 videos. Its central sample follows channels from their first upload, including the many that never took off, which most creator statistics miss. The data show that what goes into a video does not predict its success: 36 measures of images, speech and text explain at most 1.1% of the differences in views within a channel, even though they can tell what kind of video it is. Practice does not help either: forty more videos without a hit add nothing. Nothing accumulates except a hit. A channel's first hit raises the views of everything it publishes afterwards by 164% on average, a result that survives all 17 robustness checks. Creators respond by copying their hit, both its look and its topic, yet only copying the topic pays; copying the look adds nothing measurable. A simple model explains why. A hit is a lasting prize, so entering can make sense despite long odds, and because the platform's statistics describe a hit without separating its causes, copying everything about it is the sensible response. Platforms could redirect creators' effort toward what pays by showing whether a hit's topic or its look drove its success.

Repeated sampling can reveal a correct numerical answer without yielding either a reliable system output or a supported derivation. We present Coverage, Realization, and Validity Evidence (CRV), an evaluation protocol for sampled large language model (LLM) reasoning over formal geometry states. Coverage is answer availability, realization is readout accuracy on the frozen candidate pool, and validity evidence is a label-blinded critic judgment of derivational support rather than a proof certificate. CRV freezes each candidate pool before comparing readouts and analyzes covered failures by correct-answer multiplicity and within-problem discrimination. On HardShift441, a 441-problem set for which a reference solver leaves 406 problems unsolved, a LoRA-adapted Qwen2.5-7B generator obtains 24.2% average single-sample accuracy and 68.9% pass@16, whereas verifier-weighted self-consistency (WSC) reaches 38.0%. Readout accuracy is particularly low when the correct answer occurs only once or twice in the pool. In a separate constructed audit of 195 covered problems, the critic labels 12 correct-answer representatives as supported, 181 as refuted, and two as uncertain. These results show that coverage, realization, and validity evidence from the critic are distinct quantities and should be reported separately.

Wireless digital twins (DTs) rely on 3D environment models to predict radio propagation and support wireless-network decisions, yet these models are often initialized from imperfect 3D maps. Errors in building position, height, footprint, and orientation can therefore cause a high-fidelity propagation engine to simulate the wrong physical environment. In this paper, we study how a deployed wireless network can repair an existing DT using its own radio frequency (RF) measurements. In particular, we introduce Twin Residual Alignment and Calibration Engine (TRACE), a physics-grounded learning-based self-calibration framework that treats twin maintenance as residual alignment between the physical world and the current DT. Using the same sensing configuration as the physical measurements, TRACE ray-traces the current DT, coherently backprojects the measured and simulated RF onto a common world grid, and extracts the same local region around each building's current DT position. A multi-view corrector then fuses evidence across sensing nodes and neighboring buildings to predict a gated six-parameter correction per building, without relying on absolute layout or sensor ordering, and supports iterative correction through re-rendering. On 5,400 held-out samples from unseen simulated scenes at 28 GHz, TRACE reduces 3D position RMSE from 2.202 m to 0.302 m and yaw RMSE from 4.978{\deg} to 0.894{\deg}, outperforming ViT and U-Net baselines under changes in layout, building count, sensing-node count, and SNR. On measured 28 GHz RF data from the NIST outdoor courtyard, a model trained only on synthetic RF reduces mean planar wall-position error from 1.00 m to 7.8 cm, without measured-data fine-tuning or geometric labels. These results show that the discrepancy between measured and twin-rendered RF can serve as a learning signal for repairing a wireless DT.

Matrix multiplications dominate the inference cost of modern transformer-based vision models, yet existing efficiency techniques such as post-training quantization and mixed-precision inference are largely limited to the small set of fixed-width formats (INT4, INT8, BF16, and FP16) supported by conventional accelerators. We revisit stochastic computing (SC) as a way to lift this constraint: viewed as a dense adaptive quantizer, SC controls precision by bit-stream length L rather than a fixed datapath, while each multiplication reduces to a single AND/XNOR gate. We build a GPU library that emulates SC matrix multiplication at scale, exposes stream lengths as first-class kernel arguments, and evaluates SC end-to-end on image classification, object detection and instance segmentation, class-conditional image generation, and visual world-model planning. On top of this substrate, we develop a dynamic per-row mixed-precision policy that assigns stream length per token or group at matched average budget, requires no retraining, and uses the same SC hardware across schedules. Across tasks, SC remains competitive with fixed-format INT quantization at matched bit budgets, while per-row mixed precision helps maintain accuracy at lower average stream lengths. These results provide software-level feasibility evidence that SC can serve as a dense-precision substrate for fine-grained mixed-precision inference on modern vision transformers.

We introduce Adaptive LeWorldModel (ALeWM), a world model based on a joint-embedding predictive architecture (JEPA) that learns to concentrate predictive information in compact prefixes of a wide latent representation. To encourage this ordering, ALeWM learns a sequence-conditioned distribution over prefix lengths and trains the predictor to estimate the full next embedding from a sampled input prefix. As standard anti-collapse objectives encourage variation across latent coordinates and do not organize them by predictive importance, we also introduce MixSIGReg. MixSIGReg regularizes the masked embeddings against a prior-weighted mixture with Gaussian active prefixes and zeros in the remaining coordinates. As a result, the ALeWM objective encourages early coordinates to retain information useful for prediction and recursive planning. Our analysis shows that the mixture distribution used by MixSIGReg assigns higher variance to earlier coordinate blocks and lower variance to later ones. In addition, we show that, under specified assumptions, prediction error is minimized by placing the information most useful for prediction in earlier blocks. Empirically, we study the behavior of ALeWM in a controlled dynamical system with known state variables and in goal-conditioned visual control. We show that ALeWM consistently achieves higher mean success rates than tuned fixed-width LeWM, with lower planning capacity on average.

Reverse-mode differentiation of Monte Carlo light transport naively requires storing a computation graph whose size grows with path length, making it impractical for deep or high-sample simulations. Path Replay Backpropagation (PRB) eliminates this memory complexity for viewpoint path tracing by reconstructing paths from their random seeds and performing a constant-memory backward replay. However, this formulation relies on a one-path-one-pixel property that does not hold for light tracing: a single light path can contribute to many pixels, and the corresponding per-vertex adjoints cannot be recovered during replay from constant-size state. We demonstrate that this structural difference prevents a direct extension of PRB to light tracing in constant memory, and that buffered two-pass variants retain linear memory scaling with path length. We then introduce two constant-memory reverse-mode formulations for differentiable light tracing. The first, Reservoir Light Replay Backpropagation (ResLRB), compresses the set of per-path sensor connections into a single stochastically selected representative using weighted reservoir sampling, preserving a two-pass structure at the cost of increased gradient variance. The second, LRB-3-pass, introduces an additional traversal that accumulates downstream adjoint contributions prior to backpropagation, retaining all connections and matching naive reverse-mode AD in expectation. Both methods support detached and attached formulations, including differentiation of sensor splat positions induced by geometric perturbations. We validate correctness against naive AD, characterize memory and runtime trade-offs, and demonstrate inverse rendering of specular caustics driven entirely by light tracing.

We prove a sharp boundedness result on $L^4(\mathbb R^2)$ for mollifications of the disc multiplier.

Vision-language models (VLMs) are increasingly being used in scientific workflows, but their ability as agents to directly control laboratory experiments from visual feedback remains underexplored. This capability is important because many laboratory tasks do not naturally provide dense, pre-defined numerical objectives: informative signals can be sparse, intermittent, or visually ambiguous. A more general laboratory agent should instead be able to interpret visual observations, take actions to acquire useful feedback, and adapt its behavior based on the consequences of those actions. We study whether general-purpose VLMs can perform this kind of closed-loop scientific control using experimental optics as a testbed. We evaluate agents on three experimental systems that isolate distinct capabilities: a Michelson interferometer, a two-mirror cavity, and a four-mirror optical relay. The agents observe camera images, directly issue actuator and measurement commands, and retain their interaction history without receiving an engineered scalar objective during control. Across these experiments and matched simulations, we find that, given task-specific natural-language guidance, VLMs can estimate actuator-response relationships, resolve ambiguous observations through intervention, and actively create informative visual feedback when signals are sparse. These results suggest that pretrained multimodal models can serve as important decision-making agents within the experimental loop. Our work also establishes optics as a physically grounded testbed for visual reasoning and active perception in scientific agents.

Running large language model (LLM) agents in production gets expensive fast. A frontier model (the largest, most capable tier) is accurate but can cost 25 times what a small model costs per token, and the gap compounds once a workflow chains several calls together. Planner-as-Router (PaR) attacks this from a different angle. Instead of leaving model-tier selection to some component downstream, it folds the choice into planning itself. As the planner breaks a query into subtasks, it also assigns each one a model size tier (small, mid, or frontier, ordered by capability and price), so the dependencies between subtasks are visible before any specialist runs. Unlike per-call routers such as cascade routing, which look at one node at a time, PaR sees the whole workflow up front and needs no separate router model or training data. We evaluate PaR with EntBench, a benchmark of 54 enterprise agentic tasks across seven classes, graded by actually running the generated Structured Query Language (SQL) and MongoDB queries against live databases. Over 1,157 evaluations spanning eight routers and three seeds, PaR stays on the observed cost-accuracy frontier. It matches a sink-frontier heuristic (frontier model on terminal nodes only) in accuracy at comparable cost and a faithful FrugalGPT cascade at lower cost, and cuts cost 44% against all-frontier routing while giving up 2.9 points of accuracy. Several accuracy gaps fall inside the plus-or-minus six-point confidence interval of a 54-task study, so we frame PaR's advantage as frontier position rather than a clean accuracy win. We also report a preliminary observation, not a validated result: a small pilot hints that cheap routing may carry a hidden compounding penalty on compositional workflows, which we frame as a hypothesis for future measurement. PaR, EntBench, and all evaluation code are open source.

Let $X=G/H$ be a real reductive symmetric space. We consider the support of the Plancherel measure for the regular representation of $G$ on $L^2(X)$, and its Fell topology. We first recall how the Plancherel formula for $X$ (known from work of Delorme and van den Ban–Schlicktrull, among others) yields a description of the support in terms of the discrete spectra of smaller symmetric spaces. We then point out that this description simplifies for a class of real symmetric spaces $X$ which we call well-tempered. These are the spaces whose tangent space at the identity coset admits a Cartan subspace whose centralizer in the Lie algebra of $G$ is abelian. Well-tempered spaces are tempered in the sense of Benoist and Kobayashi. Under mild assumptions on the disconnectedness of $G$, we give a description of the support of the Plancherel measure for well-tempered symmetric spaces in terms of characters of Cartan subgroups of $G$, and describe the Fell topology of the support. If $G$ is a complex group and $H$ a real form, then the support decomposes as a union of explicit quotients of relative continuous-parameter spaces by finite Weyl groups. As an application, we describe the $C^*$-algebra associated to the regular representation of $G$ on $L^2(X)$, and its $K$-theory.