Papers

Black-Box Optimization (BBO) is often applied in several engineering fields and can utilize an advancement of numerical measure- ments and simulation technologies. It deals with the optimization func- tions, where an analytical description is unavailable. It relies on meth- ods that require only an input point in the search space, paired with its corresponding objective function value, obtained through non-analytical means, e.g., sensors, experiments, or simulations. Common approaches include evolutionary optimization and other metaheuristics. Since BBO methods rely solely on objective function values, they typically require many evaluations, which becomes problematic when evaluating the ob- jective function is time-consuming or expensive. This leads to using surrogate-based optimization which evaluates selected true objective val- ues and trains a regression model to approximate the objective function across the search space. Surrogate-assisted black-box optimization is a small-data learning problem because the optimizer must approximate an expensive objective function from limited evaluations. Surrogate models act as data-efficient regressors, guiding the search toward promising or informative points under a restricted evaluation budget. In this paper, a new surrogate model using artificial neural networks, called Adaptive- Fidelity Nexus Covariance Matrix Adaptation Evolution Strategy (AFN- CMA-ES), is proposed for the selective evaluation of objective functions. The experimental results show its competitive performance compared to state-of-the-art surrogate-assisted BBO methods.

Visual graph reasoning requires answering graph-theoretic questions directly from graph images, where graph topology and state are conveyed visually rather than given in symbolic form. Despite recent progress of vision-language models (VLMs), current approaches to visual graph reasoning still fail on simple visual graph problems. This reveals a fundamental limitation of existing approaches: they prioritize final-answer supervision over the intermediate recovery of an explicit graph representation that preserves graph topology and state from visual input. To address this limitation, we propose VGCompiler, a compilation-centric paradigm for visual graph reasoning via knowledge compilation. VGCompiler organizes reasoning around two compilers: a representation compiler that recovers a structure-preserving intermediate graph representation from visual input, and an operation compiler that compiles query intent under the recovered graph state into an executable graph operation. Specifically, we build VGCompiler on Qwen3-VL-8B and train it with reinforcement learning guided by a layered reward over executability, compiled graph validity, representation quality, and operation quality. VGCompiler uses a frozen observer to summarize graph and question conditions into lightweight signatures, enabling archive retrieval and code reuse across similar regimes. Experiments on three benchmarks GVLQA, VisionGraph, and VGCURE, show that Qwen-VGCompiler, built on an 8B backbone, surpasses the strongest closed-source VLM baseline by 28.9% and the strongest code-based baseline by 23.7%, while maintaining high efficiency. We further evaluate VGCompiler on three real-world domains, including metro routing, logistics delivery, and network fault assessment, where it generalizes across heterogeneous visual graphs and domain-grounded tasks.

Large language models have shown considerable potential for natural-language-driven parametric CAD modeling. However, a fundamental contradiction exists between their probabilistic generation and the deterministic requirements of CAD modeling, resulting in limitations in reliability, design-intent preservation, and geometric validity. Existing methods typically rely on large-scale annotated datasets, lack explicit modeling of design intent, and underutilize the deterministic capabilities of CAD kernels. To address these limitations, we propose ReliCAD, a unified framework that transforms uncertain LLM generation into reliable parametric CAD modeling. Through explicit design-intent modeling, ReliCAD converts user requirements into structured design specifications and explicitly models geometric relations, topological dependencies, and feature construction order. It then generates constraint-aware parametric instructions and invokes the CAD kernel through an Agent-ready API to perform geometric construction and constraint solving. ReliCAD further records runtime evidence and employs a verification-feedback mechanism to assess consistency between the generated model and the design specifications, enabling error localization and iterative repair. Experiments on the public HistCAD generation dataset and our multi-granularity CAD editing dataset demonstrate that ReliCAD significantly outperforms baseline methods, achieving 99.8% validity rate and 0.8753 IoU. ReliCAD provides a verifiable, repairable, and generalizable approach to natural-language-interactive CAD modeling.

Few-shot learning research is predominantly evaluated on accuracy alone, with limited attention to the parameter and training-sample budgets required to reach that accuracy - a real constraint for practitioners without large-scale compute. We present an ultra-lightweight (22,249-34,917 parameter) spatial-relational architecture for few-shot image classification that combines fixed Gabor edge-energy guidance with a windowed, content-adaptive patch locator. Under a strictly matched, iso-episode-budget protocol (250 meta-training episodes, 5 canonical seeds, 600 evaluation episodes per seed), our architecture achieves 5-shot accuracy gains, consistent across all five seeds, over Prototypical Networks, Relation Networks, and MAML on both CIFAR-FS and MiniImageNet, while using 27-53% fewer parameters than any baseline. It also converges in fewer training episodes, generalizes better to an unseen fine-grained domain (CUB-200-2011 birds, zero retraining), and is more robust to 50% occlusion and 25% spatial translation than all three baselines. A series of falsification ablations - zeroing relational tokens at inference and retraining without them entirely - shows that the architecture's pairwise relational computation, while present, is not the primary driver of its performance; the content-adaptive patch locator is. We report this honestly, together with a capacity sweep showing a genuine accuracy plateau near 22-35k parameters, and release full seed-level results and checkpoint hashes for reproducibility.

As LLM capabilities have expanded, multi-agent communication has emerged as an increasingly active area of research. Prevailing protocols often adopt rigid structures that introduce coordination bottlenecks and can degrade as the number of agents increases. We introduce Proxifield, a round-adaptive multi-agent protocol with decentralized agent decision-making that constructs sparse communication graphs from the evolving semantic proximity of agents. Without model training or a centralized planner, Proxifield connects agents using four routing signals derived at inference time: direct address, information needs, plan alignment, and information complementarity. We compare Proxifield with two representative coordination baselines, a centralized Star protocol and a decentralized Shared Context protocol, across two domains: Drone Search and Rescue and the collective-reasoning benchmark HiddenBench. We first ablate base-model capability and find that, in both domains, the performance of Proxifield improves with model size (35B -> 397B parameter model) and Proxifield outperforms all baselines at the largest scale. As team size increases, Proxifield's task-reward advantage over Star widens from 5.4% at (N=5) to 53.0% at (N=25) and 59.5% at (N=50), while Shared Context consistently underperforms both protocols. Proxifield is also substantially more robust to permanent agent failure, retaining 73.6% of its no-failure task reward under the most severe condition, compared with 58.3% for Shared Context and 38.8% for Star. These results demonstrate that decentralized, semantically adaptive routing can improve the scalability and fault tolerance of multi-agent systems.

Massive KV caches can cause severe memory-bandwidth bottlenecks during long-context decoding. Sparse attention methods mitigate this problem, but often drop necessary context, leading to quality degradation. We introduce Elastic Threshold Attention (ETA), an end-to-end trainable architecture that rivals dense model quality under hardware-aligned block-sparse decoding. ETA predicts dynamic, contextual thresholds directly from query representations, adjusting context retention depending on the task at hand. To learn this policy from scratch while enabling near lossless KV cache pruning at inference time, we filter attention logits through a shifted SiLU gate during training. We show theoretically and empirically that this creates a smooth, near-uniform attention floor that neutralizes sub-threshold value contributions while simultaneously causing localized attention sinks on initial tokens to disappear. To materialize these advantages, we implement a fused inference-time kernel in Triton that screens KV blocks in $O(1)$ time using cached geometric-probabilistic bounds. Across language modeling, reasoning, and RULER benchmarks, our 1.45B ETA model matches or exceeds dense quality, outperforming alternative fast decoding methods (Quest, H$_2$O, NSA) while achieving higher sparsity levels. Our kernel also achieves up to $2.15\times$ end-to-end speedups over FlashAttention-2 at context lengths of up to $512$K tokens. Finally, we introduce an offline calibration algorithm for domain-specific deployments that freezes per-head constant thresholds, cutting attention compute by an additional 27% at no quality cost.

Multimodal embodied systems require translating human intentions into interpretable and coordinated behaviors across heterogeneous modalities. However, existing multimodal agents often rely on implicit representations, limiting controllability and cross-modal consistency. We present a hybrid multi-agent framework for interactive multimodal behavior simulation that bridges semantic reasoning and physical execution through a Unified Parameter Representation (UPR). LLM-based reasoning agents transform multimodal inputs into UPR, which encodes behavioral states and interpretable control parameters for simulation agents generating synchronized 3D motion, spatialized soundscapes, and environmental behaviors. We develop BirdsongChat as a prototype implementation of the proposed framework, using interactive avian behavior simulation as a testbed that tightly couples motion, vocalization, and environmental context. BirdsongChat is evaluated on text- and image-guided scenarios involving species, behaviors, affective states, environments, and multi-bird interactions. The system achieves normalized scores of 94.4% for cross-modal coherence, 100% for affective consistency, and 92.6% for generation consistency. These results demonstrate that an explicit intermediate representation effectively bridges semantic reasoning and physical execution, improving controllability and multimodal synchronization. The proposed framework thus offers a generalizable design principle for embodied AI systems requiring interpretable semantic-to-physical coordination across modalities, with potential applications in bio-inspired ecoacoustics, swarm robotics, virtual environments, and creative multimedia.

Business intelligence (BI) is a cornerstone of enterprise decision-making and is widely used by enterprise users in software such as Power BI and Tableau. In traditional BI workflows, users need to prepare data by (1) identifying relevant tables, (2) performing data transformations, and (3) building join relationships, before they can (4) answer their business questions. These steps can be complex and time-consuming, making BI challenging. Given the strong capabilities of large language models (LLMs) in working with data, we study their ability to answer BI questions end-to-end, without requiring users to manually perform the tedious preparation steps. To do this, we harvest a large collection of real-world BI projects from public sources, and manually extract pairs of (questions, ground-truth answers) from real user dashboards. The resulting benchmark, BI-Bench, is the first benchmark to systematically study LLMs' ability on end-to-end BI. We find that even frontier LLMs perform poorly on BI-Bench, with less than 50% accuracy. To address their limitations, we design a tool-augmented BI-Agent that decomposes BI workflows into subtasks on structured data, such as search, join, and transform, and orchestrates specialized data management methods across BI stages. Furthermore, we develop a post-training framework that synthesizes training trajectories from real BI projects, enabling BI-Agent to be further post-trained using both supervised fine-tuning (SFT) and reinforcement learning (RL). BI-Agent achieves substantial accuracy gains of up to 40 percentage points with vanilla LLMs, and post-trained BI-Agent yields gains of up to 30 points. Our results highlight the importance of combining tool-augmented reasoning with domain-specific post-training in complex BI workflows, and point to promising directions for future research.

Generative Artificial Intelligence (GenAI) has rapidly evolved from an experimental technology into a foundational component of modern software systems. However, as its adoption grows, protecting sensitive personal data becomes increasingly challenging. Specifically, GenAI systems not only amplify traditional privacy threats but also introduce new inference-based risks, such as constructing detailed user profiles from seemingly harmless inputs. In response, privacy threat modeling frameworks are beginning to capture GenAI-specific privacy threats with finer granularity. At the same time, a growing number of mitigation techniques have been proposed to address these threats. However, although knowledge of both threats and mitigations continues to mature, the problem- and solution-space have developed largely independently. This position paper argues that the primary challenge in GenAI privacy engineering is not the lack of knowledge about privacy threats or mitigation techniques, but the missing bridge between them. We decompose this gap into three sub-problems: (i) lack of fine-grained threat-to-mitigation mapping for GenAI systems, (ii) inapplicable solution-space assumptions in the GenAI context, and (iii) prioritization difficulty under GenAI constraints. We derive four recommendations for future mitigation-selection approaches, and outline a suggested approach that extends established threat-to-mitigation mapping methods to GenAI-specific threat characteristics. We propose a research agenda toward more systematic privacy mitigation selection for GenAI-based systems.

Despite the widespread use and success of generative AI techniques today, theoretical guarantees on learning a distribution supported in $d$ dimensions from $n$ samples degrade as $O(n^{-1/\Theta(d)})$, though shown to be minimax optimal. We hypothesize that present bounds are too pessimistic because smoothness assumptions are not enough to capture the structure of distributions that often appear in real applications. Consequently, we introduce the class of sparse priors and define the "Sparse Dimension" as a measure of sparsity of a prior over the space of all distributions. We show that distribution learning under a $k$-sparse prior achieves a Bayesian risk lower bound of $\Omega(\sqrt{k/n})$ under common distance metrics, and show a matching (up to logarithmic terms asymptotically in $n,k$) upper bound for the TV distance under mild additional assumptions. We show the statistical equivalence of distribution learning and learning to sample in the Bayesian setting so that our results apply to learning to sample as well. While $k$ can still depend on the dimension $d$, or a notion of intrinsic dimension, our results show that learning under an appropriate prior overcomes the curse of dimensionality with respect to the dependence on $n$.

We distinguish two regimes in task-oriented quantization with a known deterministic oracle action. For an unconstrained interior oracle and a smooth strongly concave utility, quantizing the oracle action by vector Lloyd-Max minimizes a mean-squared-error surrogate and achieves a $\beta/\alpha$ approximation to the optimal $K$-level task quantizer. The reduction is exact for isotropic quadratic loss, and the corresponding task rate–distortion function is bracketed by two ordinary rate–distortion functions. Budget-constrained quadratic scheduling is different: the oracle satisfies a variational inequality, so quantizing water-filled actions is not generally optimal. We derive the exact Lloyd-type conditions for this case. The optimal action for a quantizer cell is water-filling evaluated at the cell's conditional-mean load, and the optimal encoder partitions load space into affine power-diagram cells. Thus the correct prescription is to quantize the load and water-fill its centroid. The distinction is material whenever a cell crosses water-filling active-set boundaries.

Direct imaging of exoplanets is limited by the extreme contrast between the star and the planets, which is mitigated using a coronagraph. However, optical aberrations cause starlight leakage through the coronagraph, producing speckles that obscure the planetary signal. Achieving the required contrast levels demands wavefront control with subnanometric precision. Deep reinforcement learning offers a promising alternative to traditional focal-plane wavefront control techniques by enabling adaptive correction strategies learned directly from interaction with the system. In this work, we present a fully data-driven method for post-coronagraphic aberration correction in a simulated high-contrast imaging testbed. The agent controls a deformable mirror using observations consisting of focal-plane measurements (images) and physics-informed wavefront sensing information derived from these images. We evaluate different observation representations and control strategies, and the method is validated on simplified simulations of a high-contrast imaging testbed, where it successfully creates dark holes, i.e., regions of the focal plane in which residual starlight is strongly suppressed, while approaching the performance of conventional wavefront control methods.

A 2021 proof of a special case of the Union-Closed Conjecture, by Aaronson, Ellis and Leader, has been formalised in the proof assistant Isabelle/HOL. Our discussion involves sketching their proof and displaying snippets from the Isabelle version of the proof, illustrating the extent to which mathematical reasoning can be rendered clearly in a formal language.

Parkinson's disease (PD) often manifests through speech impairments, facilitating accessible, non-invasive, and cost-effective severity assessment for early diagnosis and progression tracking. Despite advances in speech foundation models (SFMs), their cross-lingual generalization for PD severity multi-class classification remains underexplored due to limited labeled data, a lack of explainable methods and variability across languages and datasets. In this work, we evaluate pre-trained embeddings from four state-of-the-art open-source SFMs across three datasets in zero-shot and k-shot cross-lingual settings for multi-class PD severity assessment. Our results show that pre-trained speech embeddings enable meaningful cross-lingual transfer, although performance is sensitive to dataset properties, preprocessing, and adaptation strategy. Misclassifications under these conditions related to inter-speaker variability and atypical speech patterns highlight the need for more robust feature extraction and modeling for PD severity assessment while emphasizing the importance of explainability for reliable clinical insights.

Large language model (LLM) inference deployed on Kubernetes faces an autoscaling challenge that conventional web services do not: new serving replicas take two to ten minutes to start because multi-gigabyte model weights must be loaded, which makes purely reactive scaling structurally late. We ask a sharp question: among the components of a predictive autoscaler, which ones actually matter under such long actuation delays? We answer it by decomposing predictive autoscaling into four factors—token-aware demand tracking, startup-delay lookahead, a bounded uncertainty margin, and plant-state observation—and measuring each factor's contribution in isolation on production-derived heavy-tailed workloads generated by ServeGen. Our main finding is that a simple exponentially weighted moving average (EWMA) predictor with delay-aware lookahead and an upper confidence bound (UCB) margin captures most of the benefit, reducing time-to-first-token (TTFT) service-level-objective (SLO) violations from 53% (reactive, queries-per-second based) to 0.5% across five random seeds; lookahead alone is the single largest factor, a 14$\times$ reduction. Kalman filter variants do not consistently improve the cost–SLO tradeoff. Controlled experiments isolate the reason token granularity is necessary: context length, through key–value cache pressure, degrades TTFT far more than request rate at matched throughput. Finally, a validation on a real Kubernetes cluster (Qwen2.5-7B, A100, vLLM) confirms the central mechanism: a delay-aware lookahead controller cuts TTFT violations from 63.5% to 3.7% relative to reactive KEDA scaling. We distinguish throughout which findings are specific to Kubernetes actuation and which are general to LLM serving.

Synthesis pipelines increasingly claim not just that a program is correct, but that it is optimal. Such a claim has two halves with radically different verification stories. The upper bound, "a program of size m exists", is witnessed by an artifact that can be re-executed, proved equivalent to its specification, and shipped with a machine-checked certificate. The lower bound, "no program of size m-1 exists", has no witness and is discharged by running a solver until it reports UNSAT. Combinatorial optimization has known this asymmetry for decades and has largely addressed it: certifying algorithms make it explicit (McConnell et al., 2011), and pseudo-Boolean proof logging can certify optimality end to end with a formally verified checker (Bogaerts et al., 2023; Koops et al., 2025). That discipline has not reached circuit minimization. We call this the refutation gap: published gate counts for minimal XOR circuits provide no certificate for either half of the claim, and neither did 121 optimality results we ourselves produced. We close the gap with a pipeline that synthesizes minimal linear straight-line programs over GF(2), where every decisive UNSAT answer emits a DRAT proof checked by an independent third-party checker. We certify all 121 optimality results established by the project, across n = 6 to 9: 111 carry independently verified refutations, 10 are closed by a free counting bound, and none disagrees with the uncertified value. The median proof is 1.1 MB and checking costs 1.9x solving. We give five case studies where verification caught defects that code review did not, report two interface obstacles that push practitioners toward the uncertified path, and describe an adversarial audit that revealed a failure tail we were about to attribute to the problem was actually caused by our own budget.

Feature importance is central to interpretable machine learning, but the term "importance" encompasses several fundamentally different notions of relevance. We develop a unified perspective based on null importance: a population-level characterization of when a feature is irrelevant under a specified notion of relevance. We consider standard notions of null importance arising from marginal and conditional statistical relevance, predictive risk, functional invariance, and causal effects, and show how these notions answer different scientific questions. We illustrate the framework in two applications in which the distinction is particularly consequential: algorithmic fairness, where common fairness criteria correspond to different notions of null importance, and genomic perturbation modeling, where different notions of relevance lead to different conclusions about what a prediction model has learned. The framework connects three aspects of feature analysis: the scientific question defining relevance, the data and model assumptions that shape how different null notions relate, and the methods used to assess importance. We establish sufficient conditions under which null notions coincide and give counterexamples showing how they diverge when those conditions fail. We then characterize which nulls different method families target and when their zero-importance statistics identify those targets. Finally, simulations spanning feature dependence, redundancy, nonlinearity, hidden features and other standard phenomena, along with case studies on image and multiomics data, provide empirical evidence for these theoretical distinctions and their practical consequences. Taken together, these results provide a common statistical language for relating scientific questions, data-generating assumptions, and algorithms, and clarify the conclusions that feature-importance analyses can support.

A fundamental assumption in robotic perception is that the sensor's field of view (FoV) is fixed relative to the robot body. Motion-decoupled sensors, such as gimbal-mounted cameras and MEMS-based LiDARs, instead allow sensing direction to be controlled independently at runtime. This freedom creates a computational challenge: efficiently selecting useful viewing directions online in feature-dense environments. We propose PIVOT, a lightweight iterative method that optimizes sensor viewing direction along a fixed translation trajectory to maximize feature visibility. Under a conical FoV model, visibility depends only on the optical axis, yielding a two-degree-of-freedom optimization on the viewing sphere $S^2$. Coordinate-free $SO(3)$ exponential-map updates enable efficient continuous optimization without explicit angular parameterizations or exhaustive viewing-sphere search. Monte Carlo evaluations retain 98.1–99.6% of brute-force visibility with a 76–85x speedup. Photorealistic simulation and real-world experiments further demonstrate improved visual localization robustness and practical viewpoint control on a quadruped robot.

Recognizing technical errors in movement is a perceptual skill important to motor learning, but it is challenging for beginners in fast, complex sports like fencing to develop it. A coach's attention is scarce, live demonstrations vary from repetition to repetition, and video review is limited to whatever camera angle was used to record it. Coaches and advanced fencers reviewing a recording face a related problem. They can see an error, but have no way to anchor their feedback to the movement itself, and are left describing it in words the learner must map back onto their own body. We present FenceXR, an augmented reality system that reconstructs 3D movement replays from monocular smartphone video to address both problems. A Trainee module trains novices to detect common lunge errors while a Reviewer module lets coaches and advanced fencers attach text or voice annotations to a specific joint and moment in a replay, which can be shared asynchronously with a trainee. In a study with 18 novice fencers, unaided error-detection accuracy rose from near-chance (37.5%) before training to 64.1% after a single session, with interviews showing a shift from broad visual scanning toward targeted inspection of specific joints and their timing. In a video-based study with four fencing experts, all four viewed the Reviewer module as a valuable complement to their existing coaching tools, particularly for feedback that is difficult to convey through standard video. We end with a discussion of implications for designing AR systems that ground movement-based training and feedback in the movement itself.

As model-generated content is increasingly consumed by other model instances in automated workflows, a practically important question arises: can a model embed a signal in natural language that an independent instance of the same model can detect, relying only on shared pre-training and task instructions, without any shared memory or coordination-specific training? We introduce For Your Eyes Only, a cooperative signalling game designed to evaluate this directly. A Sender produces free-form descriptions for two words, one of which is a hidden target; an isolated Receiver must identify it. We evaluate seven contemporary models from four architectural families on 300 word pairs from established psycholinguistic corpora, using the Double-Pass Success Rate to control for output biases. We find that most models struggle to maintain coordination once they are required to avoid detectable signals, while one frontier model retains near-perfect performance even after such filtering. We further show that models can direct this capability toward deliberate misdirection, and that coordination is consistently weaker across architectures than within them.

Agents can now externalize experience into memory, consolidating historical traces into semantic knowledge and procedural shortcuts that persist between sessions. Such memory is typically private to a single agent. We argue that agentic memory benefits from being collective, because trustworthy knowledge of the shared environment—the data sources, services, and tools agents depend on—cannot be established by any single agent, only corroborated across many independent observers. We present Cairn, a community reputation platform that captures collective knowledge, allowing agents to query the community's opinion of a resource before use and to submit evidence-backed ratings afterward. Cairn aggregates observations via a time-decayed Beta model with confidence shrinkage and supports semantic discovery over reviewer rationales. We evaluate Cairn's reputation engine under adversarial simulation (e.g., lying, collusion, camouflage), benchmark its retrieval performance, and report a case study of rating heterogeneous agents in production.

Test-time scaling can improve large language model reasoning by generating and combining multiple candidate responses. In sampling-based methods, the inference budget is often described by the number of generated candidates, N. However, N tells us how many candidates are generated, not how they are executed. The same candidate budget can be produced in one batched generation call or split across several sequential calls with smaller batch sizes. We first study the effect of increasing N on reasoning accuracy using Phi-3-mini and Qwen2.5-1.5B on 500 GSM8K prompts. As expected, increasing N from 1 to 8 improves accuracy by 8.4 percentage points for Phi-3-mini and 18.4 points for Qwen2.5-1.5B. However, accuracy alone does not show the systems cost of using a larger candidate budget. We therefore fix N = 8 and compare four generation schedules: 1x8, 2x4, 4x2, and 8x1, where axb denotes a generation calls with b candidates per call. We measure latency, throughput, GPU-hours, and gross GPU-device energy while keeping the total candidate count fixed. On A100 GPUs, eight serial calls use 4.64-4.86x as much gross GPU-device energy and have 5.77-6.12x the P95 latency of one batched call with eight candidates. The same pattern appears across three independently scheduled A100 nodes per model and in short-output SciQ/V100 experiments. These results show that candidate count alone is not enough to describe the systems cost of multi-candidate test-time scaling. When candidates are independent and memory allows it, fewer generation calls with larger batch sizes are more efficient. Evaluations should therefore report not only candidate count and accuracy, but also generation schedule and GPU-level systems metrics.

We study a finite-domain event for the mirror model on the Manhattan lattice: every boundary-to-boundary trajectory crosses a marked connector an even number of times, while internal cycles are unrestricted. Our main theorem gives a finite-scale confinement criterion: if this event's probability reaches a universal threshold in one even two-square rectangle, then almost surely every trajectory in the plane is periodic. We construct an exactly normalized three-color tensor representation and prove deterministic bounds for approximate contractions using the signed partial-permutation structure of directed continuations. These tools turn numerical evaluations into rigorous finite-volume statements. At mirror density \(p=2/5\), we prove that the boundary-parity probability is nonmonotone in the rectangle width, with a decrease followed by a certified increase. These values remain below the sufficient confinement threshold and do not resolve planar localization at this density. Further experiments track how the measured finite-size minimum varies with density, compare matching-based and entrywise error bounds, and exhibit endpoint-sensitive and cooperative responses to local mirror changes.

Among the different ways of approaching the semantics of process calculi, true-concurrency models stand out for their ability to highlight subtle interplays between events. At their heart lie three crucial relations: concurrency, causality and conflict. This paper shows that reversibility, when endowed with a notion of independence, provides a rich tooling to study and characterise these true-concurrency relations. First, we prove that systems admitting pre-reversibility (i.e., that can be extended with an independence relation satisfying some basic axioms) have a unique notion of independence, events, concurrency, causality and conflict. We then analyse the relationship between independence and the true-concurrency relations, establishing novel independence-based characterisations of causality and conflict. Our second series of contributions revolves around two concrete process calculi and two syntactic notions defined on their transition labels, namely independence and dependence; we prove that they partition connected transitions and characterise elegantly concurrency on adjacent transitions. This part of our development was machine-checked using the proof assistant Beluga. Last, we study how the key mechanism commonly used in reversible process algebra can be used as a proxy to retrieve causality and core independence on past events. We conclude by discussing how our results extend beyond reversible systems.

Negative-resistance microwave oscillators can simultaneously sustain autonomous oscillations and coherently scatter electromagnetic waves, enabling active reflection beyond conventional linear amplification. Here we demonstrate giant resonant reflection gain from synchronization-induced phase localization in a self-sustained tunnel diode oscillator operating near 2.96 GHz. Weak resonant RF injection drives a transition from a broadband free running state to a phase localized narrowband reflected carrier. The resulting reflected enhancement reaches 83.1dB relative to the injected signal and is accompanied by greater than order of magnitude linewidth collapse, nonlinear injection pulling, and a synchronization bandwidth of 15.84 kHz. Time resolved spectrograms directly resolve injection locking and frequency entrainment, while noisy Stuart Landau simulations reproduce the observed spectral concentration and detuning dependent gain roll off near the Hopf instability. Artificially increasing phase diffusion with broadband bias noise suppresses the reflected enhancement, confirming synchronization-induced quenching of phase fluctuations as the origin of the giant coherent reflection gain.

We reduce the leading redundancy coefficient for binary codes correcting any fixed number $t\ge2$ of deletions from $2t$ to $2t-1$. This is the first improvement to that coefficient since Levenshtein's 1965 bound. More generally, over every fixed alphabet of size $q\ge2$, codes of length $n$ correcting any combination of at most $t$ insertions, deletions and substitutions exist with redundancy $(2t-1)\log_q n+O_t(\log_q\log_q n)$. For two deletions, the upper-bound coefficient drops from $4$ to $3$, halving the gap to the lower-bound coefficient $2$. The leading coefficient $2t-1$ also holds for $t$ bounded substring edits, where both the removed and inserted strings have bounded length. The proof combines random binary linear hashing of substring counts with a two-colouring argument. A linear dependence among differences of substring-count vectors forces their local changes to overlap, and this overlap lowers the redundancy coefficient from $2t$ to $2t-1$. The result provides an edit-distance analogue of Varshamov's 1957 bound in the Hamming metric, both in the leading redundancy coefficient and in the role of linear independence in the proof.

Modern LLMs and AI agents increasingly support data engineering workflows that integrate evidence from unstructured sources. Such pipelines typically do data retrieval, integration, and ranking before proceeding to more complex agentic reasoning or actions, e.g., for scientific discovery. The core retrieval problem in these workflows jointly executes multi-attribute filtering, multi-vector search, exact relational joins, and thresholded embedding-similarity joins. Given a planned query and monotone scoring function, our DASE query engine constructs and ranks candidate evidence tuples. It comprises (i) a multi-step reasoning query model over structured predicates, multiple vectors, and relational links; (ii) SemJI, a sparse materialized embedding-similarity join index for rare near-neighbor pairs; and (iii) a co-designed execution layer that combines predicate-aware ANN traversal, batched access, and threshold-based score aggregation. On scientific-discovery workloads, DASE retrieves candidate evidence for multi-step reasoning queries 6x to 46x faster than strong RDBMS, rerank, and vector-database baselines at comparable recall; and for tasks that require semantic-operator post-processing, DASE acts as a high-recall prefilter that makes downstream LLM evaluation both cheaper and more accurate – e.g., on SemBench E-Commerce it improves BigQuery quality from 0.67 to 0.80 while cutting cost from $2.42 to $0.54.

Instructional materials can be passed along to new instructors, yet the instructional reasoning, such as instructional intentions, contextual information, and reflective insights behind their design, is rarely preserved. Traditional Learning Management System (LMS) tools are designed to store instructional artifacts rather than reasoning. To address this issue, we propose Teaching Memory, a design approach that treats instructional reasoning as a primary form of knowledge. We present a design-oriented framework for preserving instructional reasoning in a lightweight manner within instructors' existing workflows, enabling continuity and reuse.

Text-based person retrieval under a sim-to-real gap (synthetic training data, a real-image gallery) is usually tackled with costly fine-tuned cross-encoders. We ask whether a frozen-encoder system can compete. We present SCOUT, which casts cross-modal retrieval as prediction in embedding space. A trainable predictor maps the patch tokens of a frozen video encoder into the embedding space of a frozen text encoder under a bidirectional InfoNCE objective, and no encoder is fine-tuned in the base model. The video encoder is V-JEPA, the text encoder is EmbeddingGemma, and the predictor is initialized from a Qwen3.5-0.8B decoder. We make three findings. First, the best frozen text encoder is simply the one whose geometry best matches the video features. A training-free alignment score ranks three candidate text encoders in the same order as their retrieval accuracy on our held-out split (Spearman $\rho = 1.0$); a fourth, LLM-based encoder shows the rule is metric-dependent, holding for a neighborhood-overlap score ($\rho = 0.8$) but not for a linear probe ($\rho = -0.2$). Second, two precision-targeted levers, parameter-efficient ExPLoRA adaptation of the video encoder and a training-free attribute-decomposed reranker built on a vision-language model, improve the top-rank precision that otherwise limits the frozen system, adding 2.2 points of leaderboard R@1. Third, a local-versus-public calibration study explains which interventions transfer to the real domain. On AI City Challenge 2026 Track 4 the full retrieve-fuse-rerank system reaches 84.25 mAP@10 on the final leaderboard, while a single frozen model submitted alone reaches 60.63. Our trained components cost about 95 GPU-hours. CMP, the dataset authors' fine-tuned cross-encoder that trains for sixteen GPU-days, is one fusion member of the full system, not an alternative. Code and annotations: https://github.com/abtraore/SCOUT-ECCV

Algebraic Retrieval lets AI agents compose search strategies at query time. Relevance criteria, eligibility constraints, and ranking preferences can be expressed together in a mathematical query. The query surface exposes available operations, so an agent can combine them for the question at hand and revise a program after inspecting results. We evaluate execution parity, not agent behavior or retrieval quality. Building on Programmatic Embedding Modulation (PEM), which exposes vector and score arithmetic during retrieval, we demonstrate contrastive scoring, candidate-pool reranking, and weighted ranking as composable queries, alongside executable SQL and PyTerrier counterparts. On the public 11,429-document Vaswani fixture, each program's implementations select the same document set with score differences below 1e-6; one tied pair orders differently across scoring paths.