arXiv:2606.25647v1 Announce Type: new Abstract: Smallholder farming communities in remote, depopulating areas have limited access to agricultural advice, and their locally specific agronomic knowledge, often expressed in regional dialect, is poorly represented in the global corpora on which Large Language Models (LLMs) are trained. A general-purpose chatbot therefore answers fluently but unreliably, ungrounded in authoritative local data farmers can trust. This paper presents a conversational AI assistant, Falco eleonorae, embedded in a bilingual (Greek-primary, English-secondary) e-market platform serving farmers and cooperatives of a defined island area of interest. It is a thin Backend-for-Frontend (BFF) proxy in front of a geospatially-aware agronomic agent rather than a self-hosted model. Answer generation and tool selection are delegated to a managed upstream service on OpenAI GPT-5-family models, while one bounded task, describing an uploaded field photograph, is handled directly by a vision-capable model so only text reaches the agent, and voice input is transcribed by a managed EU streaming speech-to-text service. Grounding comes not from a self-hosted vector database but from tool-augmented retrieval: a Model Context Protocol (MCP) tool queries a curated, read-only, bilingual data interface exposing local crops, a seasonal calendar, traditional practices, a dialect glossary, products, agritourism experiences, cooperatives, and training content, each wrapped in a geospatial Well-Known Text envelope anchoring the agent to the area of interest. We detail its multilingual, voice, and image modalities, its progressive-web-application and accessibility design for low-bandwidth field use, and its security and data-protection posture, and argue that for a small, resource-constrained rural deployment a managed, grounded multilingual assistant is more attainable and trustworthy than a self-hosted model.
Science Journals
arXiv:2510.00866v4 Announce Type: replace Abstract: Large-scale models are pretrained on massive web-crawled datasets containing documents of mixed quality, making data filtering essential. A popular method is Classifier-based Quality Filtering (CQF), which trains a binary classifier to distinguish between pretraining data and a small, high-quality set. It assigns each pretraining document a quality score defined as the classifier's score and retains only the top-scoring ones. We provide an in-depth analysis of CQF. We show that while CQF improves downstream task performance, it does not necessarily enhance language modeling on the high-quality set. Importantly, we find that training on CQF-selected data can outperform training directly on the high-quality set, even when the latter is sufficiently large. This finding alone is particularly striking, given the substantial effort and cost recently devoted to augmenting high-quality data. We explain this paradox by the fact that CQF implicitly filters the high-quality dataset as well as the low-quality one. Finally, we introduce an optimization-driven notion of data quality and demonstrate that it can be reliably estimated using small-scale proxy experiments. Altogether, our results both elucidate the mechanisms behind CQF and deepen our understanding of data selection methods widely used in practice.
arXiv:2510.00824v2 Announce Type: replace Abstract: Virtual reality (VR) introduces sensory perturbations that may impact perception and action. The current study was designed to investigate how immersive VR presented through a head-mounted display (HMD) affects perceived functional body size using a passable aperture paradigm. Participants (n=60) performed an action task (sidle through apertures) and a perception task (adjust aperture width until passable without contact) in both physical, unmediated reality (UR) and VR. Results revealed significantly higher action and perceptual thresholds in VR compared to UR. Affordance ratios (perceptual threshold over action threshold) were also higher in VR, indicating that the increase in perceptual thresholds in VR was driven partly by sensorimotor uncertainty, as reflected in the increase in the action thresholds, and partly by perceptual distortions imposed by VR. This perceptual overestimation in VR also persisted as an aftereffect in UR following VR exposure. Geometrical modelling attributed the disproportionate increase in the perceptual threshold in VR primarily to depth compression. This compression, stemming from the vergence-accommodation conflict (VAC), caused the virtual aperture to be perceived as narrower than depicted, thus requiring a wider adjusted aperture. Critically, after mathematically correcting for the VAC's impact on perceived aperture width, the affordance ratios in VR became equivalent to those in UR. These outcomes demonstrate a recovered invariant geometrical scaling, suggesting that perception remained functionally attuned to action capabilities once VAC-induced distortions were accounted for. These findings highlight that VR-induced depth compression systematically alters perceived body-environment relationships, leading to an altered sense of one's functional body size.
arXiv:2606.24956v1 Announce Type: new Abstract: Spectral graph neural networks (GNNs) interpret message passing as frequency-selective filtering. While low-order spectral filters are efficient, their limited selectivity often leads to weak attenuation outside the passband, whereas high-order alternatives introduce optimization challenges. We propose DCQ-GNN, a spectral GNN based on a compact bank of adaptive convex--concave quadratic filters. By restricting the filter order to two while explicitly exploiting complementary curvature, DCQ-GNN improves spectral selectivity as quantified by Dirichlet energy and entropy measures without resorting to high-order polynomial expansions. The model fuses filter outputs through a node-adaptive gating mechanism to enable node-wise structure-aware spectral selection. We provide a formal spectral analysis grounded in Dirichlet energy attenuation, von Neumann entropy, and curvature polarity, and derive explicit characterizations of filter behavior across varying levels of homophily and structural perturbations. Extensive benchmarks on 10 datasets show that DCQ-GNN ties for the top average rank (3.0) on heterophilic graphs and obtains the second-best rank (4.2) on homophilic graphs, remaining competitive with representative high-order polynomial spectral filters. Furthermore, under strong structural perturbations, DCQ-GNN exhibits substantially smaller performance degradation compared to both first-order and high-order baselines. These results demonstrate that curvature-aware quadratic banks provide a robust and efficient alternative to high-order spectral models while preserving optimization stability and computational efficiency.
arXiv:2506.01486v2 Announce Type: replace Abstract: Data imbalance persists as a pervasive challenge in regression tasks, introducing bias in model performance and undermining predictive reliability. This is particularly detrimental in applications aimed at predicting rare events that fall outside of the domain of the bulk of the training data. In this study, we review the current state-of-the-art regarding sampling-based methods and cost-sensitive learning. Additionally, we propose novel approaches to mitigate model bias. To better assess the importance of data, we introduce the density-distance and density-ratio relevance functions, which effectively integrate empirical frequency of data with domain-specific preferences, offering enhanced interpretability for end-users. Furthermore, we present advanced mitigation techniques (cSMOGN and crbSMOGN), which build upon and improve existing sampling methods. In a quantitative evaluation, we benchmark state-of-the-art methods on 10 synthetic and 42 real-world datasets, using neural networks, XGBoosting trees and Random Forest models. Our analysis shows that while most strategies improve performance on rare samples, they degrade it on frequent ones. The trade-off becomes larger the more the performance on rare samples is increased. However, to reduce this effect we demonstrate that constructing an ensemble of models -- one trained with imbalance mitigation and another without -- can be used. The key findings underscore the superior performance of our novel crbSMOGN sampling technique with the density-ratio relevance function for neural networks, outperforming state-of-the-art methods.
arXiv:2506.04435v2 Announce Type: replace Abstract: Social factors such as demographic traits and institutional prestige structure the creation and dissemination of ideas in academic publishing. One place these effects can be observed is in how central or peripheral a researcher is in the coauthorship network. Here we investigate inequities in network centrality in a hand-collected data set of 5,670 U.S.-based faculty employed in Ph.D.-granting Computer Science departments and their DBLP coauthorship connections. We introduce algorithms for combining name- and perception-based demographic labels by maximizing alignment with self-reported demographics from a survey of faculty from our census. We find that women and individuals with minoritized race identities are less central in the computer science coauthorship network, implying worse access to and ability to spread information. Centrality is also highly correlated with prestige, such that faculty in top-ranked departments are at the core and those in low-ranked departments are in the peripheries of the computer science coauthorship network. We show that these disparities can be mitigated using simulated edge interventions, interpreted as facilitated collaborations. Our intervention increases the centrality of target individuals, chosen independently of the network structure, by linking them with researchers from highly ranked institutions. When applied to scholars during their Ph.D., the intervention also improves the predicted rank of their placement institution in the academic job market. This work was guided by an ameliorative approach: uncovering social inequities in order to address them. By targeting scholars for intervention based on institutional prestige, we are able to improve their centrality in the coauthorship network that plays a key role in job placement and longer-term academic success.
arXiv:2606.24973v1 Announce Type: new Abstract: We introduce a dataset of 32,534 double-marked real student responses to GCSE mock exams (GCSEs are the UK's national exams, taken at age ~16), spanning 328 questions across five subjects and including handwritten work. We test whether off-the-shelf large language models agree with examiners as closely as the two examiners agree with each other. We find that models overwhelmingly agree well with the examiner consensus across subjects, with the top performing models agreeing more closely with examiners than examiners agree with each other. Models achieve high scores for subjective tasks like English essay marking, as well as handling complex and messy handwritten Maths paper scripts. Agreement is uniform near the examiner line, and not massively discriminated by model size, providing cost-effective automated marking solutions.
arXiv:2606.25424v1 Announce Type: cross Abstract: Diffusion-based text-to-speech (TTS) models have achieved significant improvements in speech quality. However, modeling sharp prosodic transitions and rapid pitch variations in expressive speech remains challenging. Existing diffusion-based TTS decoders commonly utilize periodic nonlinearities such as Snake activation function to capture harmonic structures, but this activation funcation provides limited adaptability when modeling abrupt amplitude and frequency variations. In this paper, we investigate the role of oscillatory inductive bias in diffusion-based TTS decoders and introduce an adaptive oscillatory nonlinearity that enables controllable periodic modulation while maintaining signal stability through a linear bypass component. We refer the resulting TTS system as OscillaTTS. Experiments on the LJSpeech and Emotional Speech Dataset show consistent improvements across objective and subjective evaluations, indicating improved modeling of expressive prosodic dynamics.
arXiv:2606.25436v1 Announce Type: cross Abstract: Dialogue systems based on large language models (LLMs) have advanced significantly in recent years. However, dialectal variation remains a major challenge, particularly for systems that process spoken input. LLM-based speech language models (SLMs), which integrate LLMs with speech processing components, show promise for spoken language tasks, yet their ability to comprehend dialects has not been sufficiently studied. Moreover, it remains unclear how the dialectal understanding of the base LLM affects SLM performance. This study investigates the dialectal robustness of both LLMs and SLMs using Japanese dialects as a test case. We define robustness as the ratio of performance on dialectal versus standard inputs, enabling fair comparisons. Our experiments show that SLM robustness correlates with that of their text-based counterparts. Furthermore, training with dialectal data and fine-tuning the speech encoder each improves robustness in SLMs.
arXiv:2606.25668v1 Announce Type: new Abstract: Automated decision systems (ADS) leverage predictions about individual future outcomes to inform consequential decision-making in organizational settings. Across various settings - including criminal pretrial release, clinical triage, student support, and more - it is often assumed that improved predictive accuracy is the priority consideration in determining better downstream outcomes upon the deployment of ADS. In practice, real-world case studies reveal that this is far from the case: introducing individual predictions into decision-making modifies organizational workflows, assessment, and decision-making processes in ways that require a complete re-consideration of our approach to the design, evaluation, and deployment of ADS. As a result, this Perspective develops an integrated framework for studying ADS in social systems, shifting current priorities from a purely prediction-based paradigm towards an intervention-oriented view that accounts for real-world conditions. Our aim is to improve our understanding of ADS and more meaningfully anticipate its downstream societal and organizational consequences.
arXiv:2606.25673v1 Announce Type: new Abstract: The demand for domain-specific systems-on-chip (SoCs) in artificial intelligence, robotics, and automotive systems is increasing the need for engineers with hands-on expertise on very-large-scale integration (VLSI) design from architecture specification to fabricated silicon. Yet, most VLSI courses rely on restrictively licensed electronic design automation tools and process design kits (PDKs), as well as closed-source hardware designs. We present an end-to-end open-source domain-specific SoC design and fabrication flow built around Croc, a highly customizable RISC-V platform. Built from open-source SystemVerilog intellectual property blocks and integrated with an end-to-end open-source design flow in a 130nm open PDK, Croc enables tapeout projects supporting multiple domain customization options: instruction-set extensions, accelerator co-processors, and peripherals. In our first open-source course experience using Croc, 65 students completed 33 projects, 30 of which produced manufacturable layouts. 18 designs were selected as tapeout candidates, and five were fabricated. A first baseline chip has already been successfully characterized in silicon, demonstrating microcontroller-class functionality and implementation metrics comparable to those of products with similar functional complexity completed with closed-source toolchains and PDKs.
arXiv:2510.01163v2 Announce Type: replace Abstract: The factors driving the performance of in-context learning (ICL) in large language models (LLMs) remain poorly understood despite ICL's surprising effectiveness, enabling models to adapt to new tasks from only a handful of examples. To clarify and improve these capabilities, we characterize how the statistical properties of the pretraining distribution (e.g., tail behavior, coverage) shape ICL. We develop a theoretical framework that encompasses generalization and task selection and show how distributional properties govern sample efficiency, task retrieval, and robustness. To this end, we generalize existing concentration results to heavy-tailed priors and dependent sequences, better reflecting the structure of LLM pretraining data. Our framework reveals a fundamental design trade-off: heavy-tailed pretraining distributions facilitate robust task selection under distribution shifts but are detrimental to generalization, especially in low-data regimes. We then empirically evaluate our predictions by studying how ICL performance varies with the pretraining distribution on challenging tasks such as stochastic differential equations and stochastic processes with memory. Together, these findings suggest that controlling key statistical properties of the pretraining distribution is essential for building ICL-capable and reliable LLMs.
arXiv:2506.21091v2 Announce Type: replace Abstract: Stereo matching has become an increasingly important component of modern autonomous systems. Developing deep learning-based stereo matching models that deliver high accuracy while operating in real-time continues to be a major challenge in computer vision. In the domain of cost-volume-based stereo matching, accurate disparity estimation depends heavily on large-scale cost volumes. However, such large volumes store substantial redundant information and also require computationally intensive aggregation units for processing and regression, making real-time performance unattainable. Conversely, small-scale cost volumes followed by lightweight aggregation units provide a promising route for real-time performance, but lack sufficient information to ensure highly accurate disparity estimation. To address this challenge, we propose the Enhanced Shuffle Mixer (ESM) to mitigate information loss associated with small-scale cost volumes. ESM restores critical details by integrating primary features into the disparity upsampling unit. It quickly extracts features from the initial disparity estimation and fuses them with image features. These features are mixed by shuffling and layer splitting then refined through a compact feature-guided hourglass network to recover more detailed scene geometry. The ESM focuses on local contextual connectivity with a large receptive field and low computational cost, leading to the reconstruction of a highly accurate disparity map at real-time. The compact version of ESMStereo achieves an inference speed of 116 FPS on high-end GPUs and 91 FPS on the AGX Orin.
arXiv:2510.09036v2 Announce Type: replace Abstract: Learned world models hold significant potential as neural simulators for robotic manipulation. However, prevalent 2D video-based models inherently lack the spatial and kinematic reasoning crucial for physical interactions. We introduce RoDyn, a novel Robot-Dynamic 2.5D World Model that formulates environmental dynamics within a highly efficient, geometry-aware latent space. Through the proposed Robot-Dynamic Tokenizer, we explicitly couple semantic visual appearances with spatial and agent-centric priors via an RGB-dominated cross-attention mechanism and dynamic mask guidance. Furthermore, by injecting these mask priors directly into sequence transitions, our Mask-guided Autoregressive architecture drives the model to focus on active robot-object interaction regions. Extensive experiments demonstrate that RoDyn establishes SOTA generation fidelity across large-scale datasets. Crucially, it translates these predictive capabilities into substantial downstream gains, accelerating model-based reinforcement learning and achieving a 42\% improvement in real-world imitation learning success rates over pure 2D baselines.
arXiv:2510.17059v2 Announce Type: replace Abstract: Zero-shot imitation learning requires an agent to reproduce expert behavior from a single demonstration without additional environment interaction or gradient updates at test time. We introduce Contrastive Inverse Reinforcement Learning (CIRL), a self-supervised framework for pre-training zero-shot imitation agents. Our methods rests on a key observation that many useful tasks can be summarized by a single goal state. We can thus convert the multi-task inverse RL problem into a more tractable goal-inference problem, and utilize state-of-the-art goal-conditioned RL methods to recover a policy that reaches the goal. During pre-training, CIRL jointly employs three components to learn without any rewards or demonstrations: (1) a variant of contrastive RL designed to learn maximum-entropy goal-conditioned policies, (2) an automatic goal proposal mechanism (GoalKDE) that drives exploration, and (3) a mean-field variational model that performs amortized goal inference from trajectories. We prove that this procedure consistently recovers the demonstrator's intent by accounting for the relative difficulty of reaching different states and show how structurally similar prior work may otherwise fail to infer the correct reward. Experiments on goal-conditioned and standard reward-maximizing control tasks show that CIRL outperforms prior zero-shot imitation methods, supporting the expressiveness of goals as a compact summary of behavior.
arXiv:2510.20025v2 Announce Type: replace Abstract: Monitoring systems increasingly fuse dynamic behavioral data with outcome-based surveillance, raising a basic question: when does behavioral data carry predictive information that outcome history lacks? We study this using epidemic forecasting on mobility networks, asking whether mobility networks provide independent predictive signal beyond local outcome-based surveillance. We formalize this as a substitution-complementarity problem over directed, weighted mobility networks. Using a Frisch-Waugh-Lovell variance decomposition, our analytical framework derives domain-agnostic conditions under which network-topology features retain incremental explanatory power beyond autoregressive outcome histories. We instantiate the framework using town-level COVID-19 forecasting in Massachusetts (April 2020-April 2021), constructing mobility networks among 300+ towns from smartphone-derived origin-destination aggregates to extract centrality metrics. An agent-based model on synthetic networks confirms that the regime boundary arises from a generic interaction between macro-scale epidemic state and network topology, rather than dataset-specific artifacts. Prevalence-gated interactions between statewide incidence and network features yield large out-of-sample gains when primary surveillance is degraded (Predict-R2 increases from about 0.60 to 0.83-0.89) but only marginal lift when granular local histories are available (+0.5 percentage points). Gains concentrate during epidemic waves when behavioral responses shift network connectivity rapidly. Framed as a value-of-information problem, the substitution gain reflects the marginal value of behavioral data relative to primary-channel quality. This yields a transferable, cost-aware design rule: integrate topology-aware behavioral signals when primary surveillance is degraded or the network changes rapidly; otherwise, rely on autoregressive baselines.
arXiv:2606.25924v1 Announce Type: cross Abstract: Richardson--Lucy (RL) deconvolution improves fluorescence microscopy images by recovering details lost to diffraction. It estimates the original fluorescence signal that most likely produced the measured photon counts under a Poisson imaging model. Although RL incorporates a physical model of fluorescence image formation and can improve contrast, deconvolution remains fundamentally ill-posed, and the measurements alone provide limited evidence for reliably reconstructing fine biological structure. Without additional structural guidance, RL can amplify noise and exhibit unstable convergence in low-photon regimes. Regularizers such as total variation (TV) reduce this instability but often introduce oversmoothing. Here, we investigate learned generative priors as a form of structural guidance for RL by integrating a score-based diffusion prior into a decoupled inverse-problem framework for fluorescence microscopy deconvolution. The diffusion prior is used during the RL optimization iterations, while RL retains Poisson data consistency. We validate the framework across diverse biological samples and cellular morphologies. The results show reduced RL noise amplification with improved preservation of weak filamentous and punctate structures under low photon counts.
arXiv:2606.25056v1 Announce Type: new Abstract: Behavioral Foundation Models (BFMs) offer a promising path toward universal physics-based character control by organizing a rich repertoire of physically plausible behaviors into a latent space, guided by a large-scale motion dataset. While these models excel at time-invariant tasks, such as goal-reaching and state-based reward optimization, their latent space does not directly support time-varying objectives, such as tracking a motion sequence. For tracking, existing heuristics rely on moving-window-averaging that fails to capture the nuances of highly dynamic motions. In this work, we propose a novel Latent Sequence Optimization (LSO) to address these shortcomings. Our approach combines simulation rollouts with a policy gradient update to optimize over a sequence of latents, extending the capabilities of BFMs toward precise motion tracking without requiring reward engineering and tuning. To guide the optimization toward smooth, coherent latent trajectories, we model the latent sequence using temporally correlated noise. We validate our approach across dense tracking, sparse keyframing, and direct deployment onto a real humanoid robot.
arXiv:2606.25738v1 Announce Type: new Abstract: DNA variant calling represents a key operation in bioinformatics pipelines that aims at identifying genetic variants. Given an evidenced explosion in genomic data availability, there is an urgent need for a high-performant, portable and efficient solution for variant calling, which can further improve our understanding of genomic structure and genetic basis for complex diseases. In its most common formulation, the Pair Hidden Markov Model (PairHMM) algorithm for variant calling stands as the main bottleneck in the pipeline, accounting for up to 70% of the execution time in large-scale genomic datasets. The state-of-the-art approaches for accelerating PairHMM in CPUs and GPUs do not scale to long DNA sequences and only explore very limited anti-diagonal data parallelism, which yields poor performance. In this work, Endeavor is proposed as a new parallelization strategy for PairHMM that redefines its traditional formulation to explore row-level fine-grained parallelism without loss in solution accuracy. Based on this, a novel and portable SIMD-based approach is derived for efficient and high-performance processing of short and long sequences in CPUs and GPUs, leveraging novel levels of parallelism and synchronization to achieve high throughput in sequences up to 100k basepairs for the first time. Evaluation on Intel and AMD CPUs shows that Endeavor outperforms GKL up to 2.14x in peak throughput and GATK HaplotypeCaller by at least 2x in real-world datasets, while NVIDIA and AMD GPUs achieve up to 2.05x speedups in genome-scale datasets when compared to state-of-the-art GPU-based methods.
arXiv:2606.25084v1 Announce Type: new Abstract: Multimodal Large Language Models (MLLMs) have redefined visual understanding by combining vision encoders with large-scale language models. This unified architecture enables strong performance on tasks like image captioning, visual question answering, and multimodal dialogue, often in zero- and few-shot settings. Their general-purpose capabilities and flexible interfaces make MLLMs a promising foundation for real-world vision-language applications. Assistive AI aims to help users interact with their environments through natural language. These scenarios demand robust visual recognition, contextual reasoning, and multilingual comprehension-capabilities that MLLMs are believed to offer. However, their effectiveness in assistive settings remains to be fully understood. In this work, we explore whether MLLMs can support Assistive AI by evaluating state-of-the-art models on real-world tasks: recognizing everyday objects like currency, answering questions based on scene text, and reading visually presented content across multiple languages. To this end, we developed a system, NetraLink, using a head-mounted GoPro to capture real-world egocentric data, and collected a benchmark covering these assistive scenarios. Our findings provide a comprehensive diagnostic of current MLLMs, highlighting their strengths and limitations in enabling assistive technologies grounded in visual perception and language interaction.
arXiv:2606.25091v1 Announce Type: new Abstract: Speculative decoding (SD) accelerates LLM inference by $1.5$-$3$ times when the draft and target models are co-located. This has motivated a distributed variant (DSD) that places the draft model on an edge device while the target stays in the cloud. We show with closed-form inequalities that DSD's per-request latency benefit is limited under WAN edge-cloud communication. If the server can host both models, co-located SD has lower latency and communication than synchronous DSD, with the same per-output FLOPs and model-weight memory. Pipelining can make DSD competitive with co-located SD only in low-RTT regimes where the round trip is shorter than the edge drafting time window; at WAN RTTs, the cloud round trip remains too large for pipelined DSD to beat co-located SD. Against cloud autoregressive decoding, DSD can reduce latency only inside a bounded window given the target-model speed, acceptance rate, and RTT. DSD is also infeasible against closed-source APIs without a verifier-only interface. The main case for DSD appears in multi-tenant capacity. Under cross-client overlap, offloading draft compute lets a saturated cloud server sustain $(1 + \gamma\,t_d/t_v)$ times more concurrent clients at the same per-client rate, where $\gamma$ is the speculation length and $t_d, t_v$ are the per-step draft and verification times. DSD should therefore be evaluated primarily by multi-tenant capacity and server throughput, not only by single-request latency.
arXiv:2606.25751v1 Announce Type: new Abstract: Traditionally, the critical blast wave parameters used to characterize loading conditions are obtained through pressure gauge measurements. However, these instruments are costly, require careful calibration, provide discrete location measurements only, and must be deployed in hazardous environments. Recent events, such as the Beirut port explosion have demonstrated that video recordings, which provide time of arrival (ta) versus distance data, offers valuable information for post-event blast analysis. However, methodologies capable of predicting key blast parameters, such as positive phase duration and impulse, using video data alone remain limited. This work proposes and validates a novel methodology to predict positive phase duration and impulse for spherical, non-cased, free air bursts of ideal explosives using ta data only. The proposed methodology was evaluated using experimental datasets from the literature for bulk and cartridge PE4, PE7, Composition B, and PETN. The positive phase duration and impulse models achieved, respectively, mean absolute percentage errors of 5.3% and 5.3%, maximum deviations of 20% and 9.4%, absolute biases of zero and 3.1%, and confidence interval coverages of 86% and 83%. The predicted results achieve remarkable comparison to all reported experimental data, verifying the ability to capture positive phase blast loading for high speed video; a step-change in explosive characterisation through full spatial and temporal primary shock characteristics.
arXiv:2606.25807v1 Announce Type: cross Abstract: This paper proposes a candidate framework for free-space quantum key distribution (QKD) based on geometrical-configuration modulation (GM). In the minimal implementation considered here, Alice coherently splits a single photon emitted from one source into two spatial output modes with a tunable separation, and uses the source separation $R$ as the GM variable that defines the prepared single-photon spatial superposition state. Bob records the single-photon detection coordinate in the far field or Fourier plane, providing the correlated data used for soft-input information reconciliation. Based on this physical mechanism, we first establish an $R-x$ protocol model in which the source separation $R$ and the single-photon detection coordinate $x$ are random variables, and further propose an $R-\Delta x$ extension based on the difference variable $\Delta x$ between adjacent accepted detection events to mitigate slowly varying center drift in free-space links. The framework specifies state preparation, far-field conditional probabilities, soft-input information generation, parameter estimation, reconciliation, and asymptotic candidate key-rate formulas. A complete composable security analysis further requires derive an explicit computable upper bound on Eve's information from experimentally observed parameters, together with finite-key analysis and experimental validation under free-space conditions. The proposed candidate framework (GM-QKD) provides a modulation approach based on spatial degrees of freedom in which the source geometry serves as the modulation variable.
arXiv:2510.23842v2 Announce Type: replace Abstract: Most state-of-the-art sign language models are trained on interpreter or isolated vocabulary data, which overlooks the variability that characterizes natural dialogue. However, human communication dynamically adapts to contexts and interlocutors through spatiotemporal changes and articulation style. This specifically manifests itself in educational settings, where novel vocabularies are used by teachers, and students. To address this gap, we collect a motion capture dataset of American Sign Language (ASL) STEM (Science, Technology, Engineering, and Mathematics) dialogue that enables quantitative comparison between dyadic interactive signing, solo signed lecture, and interpreted articles. Using continuous kinematic features, we disentangle dialogue-specific entrainment from individual effort reduction and show spatiotemporal changes across repeated mentions of STEM terms. On average, dialogue signs are 24.6%-44.6% shorter in duration than the isolated signs, and show significant reductions absent in monologue contexts. Finally, we evaluate sign embedding models on their ability to recognize STEM signs and approximate how entrained the participants become over time. Our study bridges linguistic analysis and computational modeling to understand how pragmatics shape sign articulation and its representation in sign language technologies.
arXiv:2511.11421v2 Announce Type: replace Abstract: Class-Incremental Learning (CIL) aims to continually learn new categories without forgetting previously acquired knowledge. Vision-language models such as CLIP offer strong transferable representations via multi-modal supervision, making them promising for CIL. However, applying CLIP to CIL poses two major challenges: (1) adapting to downstream tasks often requires additional learnable modules, increasing model complexity and susceptibility to forgetting; and (2) while multi-modal representations offer complementary strengths, existing methods have yet to fully realize their potential in effectively integrating visual and textual modalities. To address these issues, we propose BOFA (Bridge-layer Orthogonal Fusion for Adaptation), a novel framework for CIL. BOFA confines all model adaptation exclusively to CLIP's existing cross-modal bridge-layer, thereby adding no extra parameters or inference cost. To prevent forgetting within this layer, it leverages Orthogonal Low-Rank Fusion, a mechanism that constrains parameter updates to a low-rank ``safe subspace" mathematically constructed to be orthogonal to past task features. This ensures stable knowledge accumulation without data replay. Furthermore, BOFA employs a cross-modal hybrid prototype that synergizes stable textual prototypes with visual counterparts derived from our stably adapted bridge-layer, enhancing classification performance. Extensive experiments on standard benchmarks show that BOFA achieves superior accuracy and efficiency compared to existing methods.