Papers

In this paper, we examine the energy efficiency (EE) of a base station (BS) with multiple antennas. We use a state-of-the-art power consumption (PC) model that captures the passive and active parts of the transceiver circuitry, including the effects of radiated power, signal processing, and passive consumption. The paper treats the transmit power, bandwidth, and number of antennas as the optimization variables. We provide novel closed-form solutions for the optimal ratios of power per unit bandwidth and power per transmit antenna, and discover a new relationship in which the radiated power equals the total transceiver power at the EE-optimal operating point. A central finding is that the EE-optimal signal-to-noise ratio (SNR) collapses to a universal numerical constant of approximately 5.93 dB, independent of channel and hardware parameters. We present an algorithm that jointly optimizes the three design variables to achieve maximum EE under practical constraints, and provide analytical insight into whether maximum power or maximum bandwidth is optimal and how many antennas a BS should utilize. We further extend the optimization framework to incorporate quality-of-service (QoS) requirements and three advanced sleep modes of varying depth: absolute sleep, deep sleep, and idle mode. We characterize the optimal hardware configuration for each mode and determine when the rush-to-sleep strategy, which transmits briefly at the EE-optimal active configuration and sleeps the rest of the time, is optimal. Incorporating wake-up transition delays, we reveal how latency constraints and sleep-mode-specific transition times jointly dictate the optimal sleep mode for data packets with absolute deadlines. Together, these results indicate that energy-efficient operation requires treating transmission and sleep as a single coupled optimization.

Non-rigid point cloud registration depends on accurate correspondences under deformation and partial visibility. Existing coarse-to-fine methods use coarse matches to select regions for fine matching, but incorrect selection can remove valid correspondences before refinement. We introduce \paper, a conditional transport framework that connects coarse correspondence estimation to high-resolution matching. A diffusion model first estimates a coarse matching matrix. We lift the complete estimate to a fine-level source matrix and refine its scores with a geometry-conditioned bridge. We consider two bridge formulations: a deterministic endpoint-parameterized flow and a stochastic paired Brownian bridge with endpoint and noise prediction. At each step, the current matrix provides an alignment for the next prediction. Both models reuse backbone features throughout inference. Experiments on 4DMatch and 4DLoMatch show improved correspondence and non-rigid registration accuracy. The models also transfer to CAPE and DeepDeform without target-domain training. Ablations examine the roles of candidate coverage, source initialization, and refinement budget.

Hard symmetry constraints reduce model complexity, but can also erase label information. Statistical symmetry release determines when finite data and quantum measurements justify relaxing such a constraint, which directions to open, and how far to move. We connect global signal detection to local, loss-dependent improvement. A two-copy twirl–swap gate estimates task information in the symmetry-breaking complement with a dimension-independent copy count under paired-state and group-unitary access; reweighting the same records resolves representation sectors. An exact duality distinguishes this Hilbert–Schmidt signal from the larger signal accessible to bounded-outcome readouts. Local improvement is governed by the release gradient and a loss-corrected double-commutator matrix. Simultaneous confidence bounds convert empirical direction selection into certified descent, using either shared Pauli measurements or scalar probes with state-independent truncation bounds. Gaussian testing lower bounds quantify the cost of searching over unknown directions in the calibrated local experiment. Independent validation controls adaptively generated models, and a fast squared-loss bound preserves the approximation–estimation rate of a nested release path. On an eight-qubit Ising model, shared measurements certify release with 6300 times fewer shots than the specified scalar estimator on the tested budget grids. Quotient quantum natural gradient then controls parameter redundancy during training. Together, these results turn symmetry relaxation into a statistically justified model-selection decision.

Segmenting bruises is a challenging task in medical imaging due to limited data and annotations, diffuse boundaries, and highly variable appearance. In this work, we propose BruNet, a segmentation framework that combines a ViT-based visual encoder (a self-supervised DINOv3 or a pretrained LingBot-Vision backbone) with a SAM-based mask decoder. BruNet is trained on the HAM10000 skin lesion dataset and evaluated on a separate bruise dataset without additional fine-tuning. Although a small number of prior studies have explored machine learning and computer vision for bruise analysis, existing work has primarily focused on detection, classification, or colour analysis rather than pixel-level localisation. To the best of our knowledge, this is the first study to address automatic bruise segmentation. Our results show that BruNet outperforms CNN-based models, state-of-the-art segmentation models, ChatGPT-4o/5-assisted SAM2 zero-shot baselines, and the medical-oriented MedSAM model, demonstrating strong cross-domain generalisation to bruise segmentation.

Reviewer comments naturally relate to specific parts of the reviewed paper, yet grounding these comments to the underlying evidence is difficult due to long multimodal documents. Existing benchmarks do not capture this setting and largely focus on explicit, information-seeking queries. We introduce ReGround, a large-scale dataset for reviewer comment grounding that links 10,267 reviewer comments to 16,274 evidence in the original anonymous submission of 3,656 papers. We build on a simple observation: author rebuttals often include explicit references to content of the submission used to address reviewer comments, providing a high-precision annotation source. We cast grounding as a retrieval task and evaluate a wide range of retrieval methods. Results show that retrieval over the entire paper content performs poorly, evidence-type inference is a major bottleneck, and multimodal evidence provides complementary signals that text alone misses. Our dataset exposes grounding reviewer comments as a difficult and practically important problem for scientific document understanding.

Determining the differences between two speakers' accents is a fundamental task in linguistics and speech technology research. The methodology used to measure these differences depends on the specific research area. A phonetics researcher may demonstrate accent variation by comparing vowel formants in paired recordings of individual words. These results will be interpretable, but the recordings will be time-consuming to collect and may not be representative of connected speech. Accented Text-to-Speech (TTS) research has pushed towards using accent embeddings derived from accent classification tasks. These embeddings can be produced from any speech recording, but are not readily interpretable. In this paper, we demonstrate that articulatory representations created through articulatory inversion can be used as an interpretable basis for accent comparison and that optimal transport provides a framework for accent comparison across arbitrary recording types.

Efficient routing optimization is essential to freight transportation, urban logistics, and shared mobility, where high-quality heuristics are often required under limited computational budgets. Recent large language model (LLM)-based automated heuristic design methods can generate effective routing rules, but aggregate evaluation may mask recurrent failures on particular instance structures. To address this limitation, this study develops RouteRepair, which diagnoses parent-specific weaknesses from instance-level performance and applies targeted modifications to the corresponding heuristic components while protecting behavior that already performs well. Routing evidence, solver behavior, and program context are combined to define bounded repair objectives, and each intervention is validated through matched parent-child evaluation of failure recovery and collateral degradation. Experiments on the traveling salesman problem (TSP) and capacitated vehicle routing problem (CVRP) span constructive search, guided local search, and ant colony optimization. RouteRepair-GLS reduces the mean TSP optimality gap from 1.7476% to 0.7587%, while the constructive CVRP heuristic lowers average route cost by 1.91% relative to the savings heuristic; the generated ACO priors also outperform matched hand-designed priors. These results show that failure-aware, evidence-constrained refinement can improve routing heuristics on difficult instances while preserving performance on cases they already solve well.

Background: To determine whether cross-lingual clinical annotation projection can be formulated as a text-preserving, document-level generative task that produces verifiable character-level annotations for multilingual clinical corpus construction, and to characterize its robustness and computational trade-offs relative to candidate-based projection pipelines. Methods: We developed a constrained LLM projection workflow that inserts entity tags directly into immutable target-language text, followed by deterministic validation and character-offset reconstruction. We evaluated it alongside supervised candidate-span projection and hybrid ML-LLM refinement for transferring Spanish Disease, Symptom, and Procedure annotations into six languages. Evaluation used MultiClinAI gold standard with strict span matching and character-overlap F1 Results: Direct LLM projection achieved the strongest and most consistent performance. GLM 5.2 obtained a mean Strict F1 of 0.9201 across 18 language-entity combinations, while locally deployable Gemma4:31B achieved 0.9133. The best LLM configuration improved Strict F1 over the previous state of the art in all 18 settings, by 0.0564-0.1512, yielding 55,416 grounded mentions with reconstructed offsets. Conclusions: Direct LLM-based projection enables high-quality multilingual clinical annotation transfer and provides a practical approach for extending clinical NLP resources to languages with fewer annotated datasets and language-specific tools. Combined with local inference and deterministic validation, it can substantially reduce expert time and cost for multilingual clinical corpus construction.

Many ML datasets are constructed by running a detector, heuristic, or model over candidate pools; accepted items become labels. Dataset precision is then governed by true-positive prevalence in each pool via Bayes, not solely by detector quality. Using one instrument and period, we hold a detector-defined event dataset plus an independent official index labeling every detected item as real or phantom. One detector, three pools yield phantom rates 81.7%, 9.0%, and 0.0%. Transferring precision from the two high-rate pools to the low-rate pool predicts 0.955 versus measured 0.183, a +422% error; the Bayes expression predicts all three within 3.3%. The detected response curve is an exact convex combination of a true-event and a phantom component (residual 1.1e-16), with phantoms outnumbering true events 473 to 308, so contamination is a second signal with detector-inherited shape, not additive noise. Contamination direction depends on the estimator: on identical windows one statistic is diluted and another inflated because its denominator is also contaminated. A common normalization turns the estimator into a mean of ratios whose expectation need not exist; on the same 335 events it returns 0.40 where the well-defined estimator returns 0.10.

This work investigates a number of approaches to prevent catastrophic forgetting in class incremental learning scenarios for sound event classification tasks. We analyze the problem using architectural and regularization approaches, using FSD50K and AudioSet datasets. We design incremental stages and solutions that selectively protect the kernels of the network from weight updates to prevent catastrophic forgetting, and a dynamic head solution that expands itself each time a new task is learned. The findings show that catastrophic forgetting mainly happens in deeper layers, in particular in the classifier head. For the studied in-domain sound classification problem, the solution that seems to alleviate catastrophic forgetting and is the most efficient is a full freezing of the feature extractor with a fine-tuning of the dynamic head classifier, showing little to no forgetting and great training stability, and a good balance between memory-stability and learning plasticity.

Heterogeneous model collaboration seeks to exploit the complementary strengths of different models to balance predictive performance and inference cost. Existing approaches typically rely either on trained routers, which tie routing decisions to a fixed task and model pool, or on raw-confidence cascades, whose thresholds lack consistent reliability semantics across heterogeneous models. Consequently, these approaches adapt poorly to changing model pools and deployment budgets. We propose Calibration-Aware Uncertainty Cascades (CAUC), a simple post-hoc framework that independently calibrates each model's confidence and selects deployment policies using validation data. The resulting calibrated confidence scores establish a common reliability scale for accepting an early prediction, invoking a stronger model, or selectively combining model outputs. This unified decision criterion decouples deployment policies from any particular model pool or operating budget. We further show theoretically that calibration gives confidence thresholds an explicit selective-risk interpretation, whereas uncalibrated scores offer no comparable reliability guarantee. Extensive experiments demonstrate that, across six language benchmarks, CAUC achieves an average relative accuracy improvement of 1.9% over strong-model-only inference while avoiding approximately 47% of strong-model calls. On image classification benchmarks, it maintains or improves predictive performance while reducing measured GFLOPs by up to 57%.

Reliable robot deployment requires online failure monitoring, yet existing monitors mainly derive risk from proxy signals or train dedicated monitoring components. We ask whether the internal predictive states of a frozen pretrained robotic world model already contain directly decodable failure information. Failure-Aware Readout from World Models (FARM) trains only a 33,985-parameter supervised readout over frozen VLA-JEPA predictive states, producing step-wise failure scores and causal trajectory risk. Five-fold out-of-fold evaluation across seven source tasks reaches 85.68/88.59 pooled AUROC/AUPRC, and FARM gives the best Seen performance among 15 matched baselines on the 10-task benchmark. Across four real-robot populations on PIPER X, SO-101, and Franka, fixed-readout transfer and readout-only adaptation test deployment shifts without updating the predictive backbone. FARM also discriminates failures from partial causal histories and adds 0.2256 ms mean CUDA latency once the frozen state is available. These results support frozen predictive world-model states as reusable features for causal, transferable, and low-overhead execution monitoring.

This paper is the first in a three-part series on the construction of polynomial splines with the highest order of smoothness over hierarchical T-meshes, referred to as $\PHtwoT$-splines. For splines of bi-degree $(d,d)$, we study suitable refinement conditions for the subsequent basis construction, which requires dimensional stability of the underlying spline space. We present two groups of examples, considering unrestricted hierarchical refinement and refinement without vanishable T $l$-edges, respectively. The first setting permits new edges without additional degrees of freedom. In the second group, every refinement level excludes vanishable T $l$-edges and increases the dimension. Nevertheless, the dimension is unstable in both groups. We then introduce template translations to describe each refined region as a union of translates of a fixed template in the cell-index grid. Together with the known stability result under $(d-1)\times(d-1)$ template refinement, these examples justify this condition as a reasonable mesh assumption for the subsequent $\PHtwoT$-spline construction.

Static side-channel attacks, which exploit halted-clock conditions to extract sensitive information, pose an increasing threat to chip security. To counter these attacks, various defenses have been proposed that monitor for abnormal clock behavior and trigger the clearing of sensitive data when clock anomalies are detected. In this work, we demonstrate that exposing a chip to cryogenic temperatures interferes with on-chip mixed-signal components, responsible for signal sensing and generation. Based on this observation, we develop Chypothermia, an attack that, without any electrical tampering with the system, disables the target clock sensor, the clock generation circuit, and the voltage sensors, while preserving the secret data. While effective at halting the clock, cooling is a slow process and, on its own, is often insufficient against systems equipped with temperature sensors designed to detect thermal anomalies. To bypass these protections, we combine Chypothermia with Chypnosis (Mitard et al., IEEE SP 2026) and show that, even within a moderately low-temperature operating range, this combination can halt the clock while evading detection. We implement Chypothermia on multiple FPGA/SoC platforms and demonstrate successful disabling of both soft-IP and hard-IP sensor implementations. Finally, we apply Chypothermia to the alert handler of the OpenTitan root of trust, which incorporates a state-of-the-art clock sensor, and show that the attack evades detection and prevents key zeroization. Finally, we introduce and implement an FPGA-compatible self-heating sensor as a countermeasure and demonstrate its robustness against Chypothermia.

It is natural for humans to walk and talk simultaneously. This paper tackles the challenge of replicating such natural behaviors in 3D avatar motion generation driven by concurrent multimodal inputs, such as a text description of a man walking alongside speech audio. Existing methods, constrained by the scarcity of aligned multimodal data, typically combine motions from individual modalities sequentially or through weighted sums. However, they often result in mismatched or unrealistic movements. To overcome these limitations, we propose MOCO, a novel diffusion-based framework capable of processing multiple simultaneous inputs, including speech audio, text descriptions, and trajectory data, to generate coherent and lifelike motions without requiring aligned multimodal data. Our key innovation lies in decoupling the motion generation process. During each denoising step, the diffusion model independently generates motions for each modality from the input noise and assembles the body parts according to predefined spatial rules. The resulting combined motion is then diffused and serves as the input noise for the subsequent denoising step. This iterative approach enables each modality to refine its contribution within the context of the overall motion, progressively harmonizing movements across modalities. Consequently, the generated motions become increasingly natural and fluid with each iteration, achieving coherent and synchronized behaviors. We evaluate our approach using a purpose-built multimodal benchmark. Experimental results demonstrate that MOCO outperforms existing baselines, advancing the field of multimodal motion generation for 3D avatars.

In modern power systems with high penetration of Inverter-Based Resources (IBR), most converters operate in Grid-Following (GFL) mode. Some buses exhibit inherently low Short-Circuit Ratios (SCRs), a property majorly shaped by network topology. The integration of GFL-IBR onto such weak buses thus demands attention to static voltage stability. To address this issue, market mechanisms have been proposed to incentivize generators to provide voltage stability services, such as commitment of synchronous generators for reducing the equivalent impedance at low SCR buses and adaptive reactive power support from GFL-IBR. To compute shadow prices for these services, previously proposed methods such as the `restricted' and `dispatchable' approaches may fail to guarantee operating cost recovery for voltage-stability service providers. As the resulting prices are determined purely from a social surplus maximization objective, the profitability of units is entirely overlooked. This suggests that new pricing methodologies are needed to satisfy cost-recovery requirements. Therefore, this paper proposes a pricing method based on a primal-dual formulation. Case studies demonstrate that the proposed method can consistently produce revenue-adequate shadow prices, enabling all participating units to recover their costs without supplementary uplift payments.

We introduce Complex-Valued Quadratic Phase Gaussian (CVQPG), a novel hologram representation method that augments each 2D Gaussian primitive with a quadratic phase profile controlled by a learnable curvature parameter. Against the planar Gaussian baseline, CVQPG improves the average PSNR of holographic reconstructions by 0.19 dB (RGB) and 0.33 dB (grayscale) at equal primitive counts, and by 0.05 dB (RGB) and 0.08 dB (grayscale) at equal parameter counts, where it still leads in all visual quality metrics. Our frequency-domain analysis shows that CVQPG better preserves the mid-to-high frequency band of natural images, where the reconstruction MSE drops by up to 11% (RGB) and 22% (grayscale), indicating that modulating primitive wavefronts is an effective and lightweight enhancement.

Skill adaptation frameworks based on reinforcement learning often require restrictive assumptions to maintain stability, such as fixed observations or tightly controlled exploration schedules. In cluttered and dynamic environments, however, unrestricted exploration can lead to unsafe behaviour and unstable learning, particularly when task-relevant observations lie near obstacles or involve moving objects. In this work, we present Dist-GPRL, a distance-aware and safety-guided reinforcement learning framework for structured robot skill adaptation. Building upon Gaussian Process (GP)-based skill parameterisation, our framework sequentially adapts overlapping local windows of sparse trajectory via-points rather than modifying the complete skill at every policy step. Raw policy outputs are correlated through the GP covariance structure, producing temporally coherent trajectory updates while reducing the action-space and credit-assignment difficulties associated with global trajectory adaptation. Safety is incorporated through two complementary forms of guidance. A safe-subspace prior derived from the Hausdorff Approximation Planner (HAP) biases policy exploration toward feasible regions, while dynamically updated distance field clearance and gradient rewards provide local obstacle awareness. A trajectory-kinematics similarity regulariser further preserves the demonstrated velocity and acceleration characteristics during adaptation. We evaluate the framework on two dynamic object-manipulation tasks in simulation and transfer the learned policy to real-world robot execution. Experimental results demonstrate higher task success, lower collision frequency, and more stable learning than the baselines, while preserving the kinematic characteristics of the demonstrated skill.

Genetic programming and its variants, such as grammatical evolution, are widely used in symbolic regression to derive mathematical expressions from multivariate data. In addition to predictive accuracy, models are appreciated for their potential to provide interpretability, offering explicit equations that relate input variables to outcomes. However, achieving interpretability and plausibility remains challenging, as evolved models may be complex or scientifically inconsistent. In this study, we explore whether large language models (LLMs) can assist in improving the interpretability of symbolic regression models generated by evolutionary computation methods. Building on our previous work on estimating body fat percentage using grammar-based genetic programming, we investigate LLMs as post-processing tools to analyze and rank evolved expressions by interpretability and physiological plausibility. Four symbolic expressions are analyzed by three LLMs over three repeated runs, and the resulting interpretations and rankings are assessed by a panel of three clinicians. Across the three LLMs, comparative model-ranking outputs received more favorable clinician assessments than isolated term-level interpretations. However, the LLMs also produced physiologically and mathematically questionable explanations, indicating that they are better suited to comparative auditing under expert oversight than to autonomous validation.

We present a novel method for computing global illumination by expressing the solution to the light transport equation as a 13D Gaussian mixture model over positions, directions, surface normals, and material properties. We show that including scene properties in the Gaussian representation drastically reduces the number of functions and speeds up evaluation. As opposed to traditional light transport methods based on Neumann series, the parameters of our model are directly estimated by minimizing the residual of the rendering equation. While both optimization and rendering require repeated evaluations of a linear combination of high-dimensional Gaussian functions, we introduce an efficient culling strategy to keep the optimization tractable and produce renderings in real time. Our representation enables to render fast, view-independent solutions to the light transport equation, achieving rendering times on the order of milliseconds, with a fraction of the memory requirements of conventional neural rendering approaches.

Scientific workflows increasingly use repeated queries, multiple models, and interacting agents. Reliability therefore depends on whether models preserve correct conclusions, accept valid corrections, and contribute errors that a selector can distinguish. Using SycoBench- 600 as a controlled measurement substrate, we evaluate these requirements through selective updating, defined by resistance to misleading suggestions and uptake of correct suggestions. The study covers ten models and 17,055 trajectories. Published models span 13.4 to 71.6 percentage points in selectivity. Under identical local evaluation, Qwen3-4B is selectively adaptive at 45.6 points, Gemma3-4B is destabilized at minus 14.1 points, and SmolLM3-3B follows both correct and wrong explicit suggestions, producing zero selectivity. Matched interventions identify model specific responses to doubt, authority, and explicit advice. Among seven published models, the best reaches 95.3 percent accuracy, plurality reaches 88.6 percent, and the oracle ceiling is 99.8 percent. Mean error correlation of 0.285 reduces seven models to an effective independent count of 2.58. A leave-one-stem-family-out reliability selector reaches 96.2 percent, recovering 67.7 percent of the plurality-to-oracle gap. These results establish selective updating, error diversity, and calibrated adjudication as jointly measurable design targets for multi-model scientific workflows.

We study asynchronous continuation replanning in a two-population linear–quadratic mean field game with deterministic open-loop controls and heterogeneous, possibly erroneous initial information. At exogenous opportunities, one population recomputes its continuation response from the physical aggregate state reached, with the opponent's active plan frozen. The state–plan target needed to initialize replanning is recoverable from local aggregate observations if and only if a kernel-inclusion condition holds; full recovery of the hidden initial input is unnecessary. For revisions accumulating before the terminal time, spectral stability of the limiting two-response cycle yields bounded physical execution and a finite state left limit. The zero-extended continuation plans then converge strongly to the mean field equilibrium restarted from that state. A scalar counterexample shows that well-posed continuation problems and a convergent physical state can coexist with divergent remaining plans. We also construct an exact causal implementation from local aggregate observations and establish finite-population robustness on every fixed finite opportunity prefix. Revision records match with probability tending to one, and the mean-square implementation error truncated to the matching event is \(O(N_*^{-1})\), where \(N_*\) is the smaller population size.

We stabilize stationary states of coupled multi-species McKean–Vlasov equations by feedback control, extending a spectral approach based on the Wasserstein Hessian and a Riccati equation from a single density to a product of Wasserstein spaces. We give necessary and sufficient conditions for local exponential stabilization at a prescribed rate when only one species is actuated, together with the minimum number of control inputs required. For two species, the condition reduces to a rank condition on the Fourier coefficients of the cross-interaction kernel over the eigenspaces of the unactuated species' linearized dynamics. A two-community noisy Kuramoto model and a cell-sorting model illustrate the results.

Effective bug triage is crucial for streamlining the software development lifecycle by accurately categorizing and assigning reported software defects. In this paper, we propose an automated bug triage system built upon the pre-trained RoBERTa-base transformer architecture. By leveraging deep contextual representations, our approach efficiently classifies incoming bug reports to optimize assignment. Experimental evaluation demonstrates that the proposed system achieves a strong bug identification accuracy of 0.90 within just five training epochs. These findings highlight the efficiency and high performance of fine-tuned transformer models for practical software engineering automation.

The Ball-Nayar-Tkocz conjecture asserts that, for independent real random variables $X,Y$ with a common log-concave density, the function $t\mapsto h(\sqrt tX+\sqrt{1-t}Y)$ is concave on $[0,1]$. Gaussian distributions have this property. We construct strongly log-concave counterexamples arbitrarily close to the Gaussian and study their persistence under Gaussian smoothing. Let $X$ have mean zero and variance one, let $G$ be an independent standard Gaussian, and let $q\in[0,1]$ denote the signal variance proportion in the smoothed variable $\sqrt qX+\sqrt{1-q}G$. In the asymmetric case, for every fixed $0<q\le1$, there are counterexamples for which the entropy of the weighted sum of two independent copies of the smoothed variable has strictly positive second derivative near the endpoints. In the symmetric case, for every $4/7<q\le1$, there are counterexamples for which an unequally weighted sum has strictly greater entropy than the equally weighted sum. Both families have smooth, strictly positive densities and are strongly log-concave before and after smoothing. The density ratios $f/\varphi$ converge uniformly to one, and each derivative of fixed positive order converges uniformly to zero, where $\varphi$ is the standard Gaussian density. The symmetric counterexamples can also match any prescribed finite number of Gaussian moments exactly. The asymmetric construction uses an endpoint expansion of entropy, whereas the symmetric construction exploits the different decay rates of high-order Hermite perturbations under different weights. Thus, densities violating concavity approach the Gaussian density in the strong sense described above. The question of uniform concavity in the symmetric class for $0<q\le4/7$ remains open.

Recent work quantified the notion of quantum counterfactual gain for an extended Elitzur-Vaidman bomb test style scenario, through a connection to the negativity of the Kirkwood-Dirac quasiprobability distribution. We here extend this work to identifying quantum advantage in a new scenario, which we term the “Quantum Plumber's Problem”. In this scenario, we imagine a “quantum plumber”, who knows that one path of an interferometer is blocked, and wants to find the optimal strategy for identifying with certainty which path this is. We discuss various strategies for a generalised path-encoded interferometer, as well as for the specific case of Hofmann's three-path interferometer, introduced in a recent analysis of the relationship between states in five measurement contexts of a three level system. We support our arguments on the relative merit of competing strategies with data collected over many simulated attempts at locating blockages. We also present results for a variant of the game in which the blockage is replaced by a non-demolition detector.

Large language models exhibit complementary strengths, motivating routing methods that dispatch each query to the most suitable model. Although existing routers are effective in single-turn settings, they do not directly transfer to multi-turn dialogue, where routing performance critically depends on how historical context is segmented, retained, and incorporated into the current prompt. This introduces two fundamental challenges: preventing information loss and information confusion during context construction, and evaluating routing quality without conflating model selection with prompt construction quality. In this paper, we propose SWRouter, a Similarity-Contractive Window Router for multi-turn large language model routing. SWRouter combines a similarity-based context segmentation mechanism for prompt construction with a dual-metric evaluation framework that decouples construction accuracy from router performance. Experiments on multi-turn dialogue benchmarks demonstrate that SWRouter consistently surpasses strong baselines, achieving a 16.26% improvement in evaluation accuracy over the best individual large language model and an additional 8.22% gain over the Conv-ID Context baseline. Our results highlight that multi-turn large language model routing requires a joint design of context construction and evaluation, rather than a direct extension of single-turn routing methods.

Solana is a rapidly growing high-throughput blockchain platform that has attracted substantial liquidity and user activity. However, this expansion has also drawn the attention of illicit actors, who frequently leverage cross-chain bridges to route illicit funds onto Solana to obfuscate transaction lineage. Unlike EVM-compatible platforms, Solana features distinct execution dynamics and lacks standard event logs, creating severe semantic gaps that prevent existing tracing methods from reliably correlating cross-ledger transactions. In this paper, we formalize four types of Solana-bound cross-chain transaction modes and propose a candidate-set selective decision-based tracing method called SolTracer. SolTracer maps disparate execution semantics into a unified event space and employs candidate-set selective decision-making to reliably associate target transactions while abstaining when valid targets are absent. Extensive experiments demonstrate that SolTracer outperforms state-of-the-art (SOTA) methods across three representative scenarios: closed-world association, open-world association, and cross-source-chain generalization. In particular, under the challenging open-world setting with a 50% TA ratio, SolTracer improves the F1 score by 20.16% over the strongest SOTA baseline. Utilizing SolTracer, we conduct an empirical analysis on real-world cross-chain transfers to investigate ecosystem dynamics. Our analysis explores the stark count-value divergence across bridge mechanisms, the prevalence of cross-asset shifts, and the decoupling between on-chain settlement and explorer visibility.

Reducing audio-encoder depth lowers the inference cost of speech large language models, but removing complete blocks perturbs the embeddings consumed by the decoder and can cause deletion and premature end-of-sequence errors. We introduce X-AuT, a progressive framework that selects layer combinations through short behavioral probes and restores the pruned model through representation alignment, cross-scale distillation, scheduled student-policy supervision, and LoRA finetuning. The language-model backbone remains frozen, while attention LoRA adapters and the tied output embedding adapt during distillation. Training uses the highest-agreement tier from a transcript-consistency pipeline, followed by source reweighting during finetuning. On ten public Chinese–English benchmarks, compressing Qwen3-ASR-0.6B from 18 to 16 audio-encoder layers reduces macro-average error from 5.61% to 5.27%. The 14-layer model reaches 5.75% with 20.7% fewer audio-tower parameters. Under the matched recipe, the 1.7B teacher yields 5.55% mean error, compared with 8.45% for self-distillation, and progressive 18$\rightarrow$14 pruning outperforms direct pruning (5.75% vs. 6.73%). These single-run results establish two practical operating points and show that the accuracy effects vary across benchmarks. Project website: https://xpeng-ai.github.io/x-aut

Trusted Execution Environments (TEEs) have become a cornerstone of modern cloud computing, providing strong confidentiality and integrity guarantees for both code and data. A critical component of this trust model is remote attestation, which enables external entities to verify the authenticity and integrity of code executing within a TEE through cryptographic measurements. However, the effectiveness of remote attestation fundamentally depends on the verifier's ability to trace the reported measurement back to the original source code - a property that can only be guaranteed through reproducible builds. In this paper, we investigate the reproducibility of TEE builds through a technical analysis of 115 TEE deployments. Our analysis spans popular TEEs such as Intel SGX, Intel TDX, and AMD SEV, and reveals that a striking 91% of those deployments were not reproducible, with 80% failing to provide both source code and a reference build, the two essential prerequisites for reproducibility. To explore the root causes, we contacted the maintainers of 50 SGX projects and managed to recruit 12 developers from industry and academia for interviews. Only one of our participants reported that reproducibility is a priority during development, effectively confirming our technical findings. Beyond technical barriers (e.g., timestamps included in the binary) that can be readily addressed, we identify broader ecosystem-level challenges, such as the lack of control over the build environment in projects involving multiple stakeholders. We argue that achieving reproducibility in TEEs requires a holistic development approach that extends beyond individual developers and calls for stronger commitments - rather than treating TEEs as a "security badge".