Papers

Remote attestation is a procedure by which a system (such as a Trusted Execution Environment) proves its own configuration to a remote peer. Here, the configuration includes the measurements (hash digests) of software components. When multiple nodes perform mutual attestation, each node must maintain the reference values for its peers' measurements. We refer to this as the reference-value bootstrapping problem for mutual attestation. In this paper, we formulate the problem as a fixed-point equation and solve it directly using Kleene's second recursion theorem. In our approach, nodes can reconstruct one another's source code, thereby allowing each node to derive the reference values at runtime. When a source code itself is deployed, as with a Python script, a node derives the peer's reference value directly from its reconstructed source. When a build artifact is deployed, as with AWS Nitro Enclaves, a node instead reproducibly builds the peer's artifact from its reconstructed source and derives the reference value from it. As a proof of concept, we develop transpilers in Python and Nix to generate mutually referencing programs. Using these transpilers, we demonstrate the mutual derivation of reference values for the two deployment models mentioned above. Our solution is architecture-independent, requires no trusted third party, and works with existing attestation stacks without modification.

Many agentic systems must repeatedly choose between acting and abstaining, making faithful reasoning important for oversight: an explanation is useful only if it reflects the computation that produced the action. We study this problem through intervention timing in multi-party conversation, where an assistant must decide whether to speak or remain silent. This setting exposes class imbalance, asymmetric action costs, and the possibility that exposing reasoning changes the policy being audited. Using Qwen3-8B, decoded with or without chain-of-thought reasoning, we compare direct decision policies, reasoning policies, supervised fine-tuning, and reinforcement learning. We find a capability-auditability tradeoff: the strongest direct policy achieves higher quality but exposes no reasoning to inspect, while the reasoning policy provides a trace at the cost of lower performance, particularly recall of true intervention opportunities. Supervised fine-tuning either suppresses reasoning or preserves it without improving decision quality, while reinforcement learning also fails to improve the reasoning policy. We identify one mechanism underlying this failure: group relative objectives provide no learning signal on confidently wrong prompts when sampled rollouts all select the same action. Controlled activation probes and behavioral ablations show that standard faithfulness methods can overstate evidence that exposed reasoning reflects the underlying decision process. Probability-based metrics saturate under confident decisions, probes are vulnerable to class imbalance and textual leakage, and reasoning ablations can confound reasoning content with changes in inference mode. Together, these results show that exposing reasoning can change an agent's action policy rather than simply make it observable. We provide controls for evaluating reasoning-based oversight of agents that can act or abstain.

In recent years, a white-box neural network called ReduNet has been proposed, which employs the maximal coding rate reduction (MCR$^2$) principle to transform raw data into low-dimensional discriminative features via a forward layer-wise construction process. Unlike traditional deep networks that rely on backpropagation, ReduNet explicitly derives the parameters of each layer from the features of its preceding layer, offering a mathematically interpretable paradigm. However, this layer-wise construction often requires a large number of layers for the MCR$^2$ objective to reach a stable value, which increases the parameter storage of the unfolded module. To address this issue, we propose LA-ReduNet, a lightweight adaptive architecture that refines the layer-wise update rule and enables discriminative feature representations to be obtained with substantially fewer unfolded layers. Specifically, LA-ReduNet employs hyperspherical manifold learning and adaptive step sizes, thereby reducing by an order of magnitude the number of layers required for the MCR$^2$ objective to reach a stable value. Simulation results demonstrate that, while maintaining comparable classification accuracy, LA-ReduNet requires significantly fewer layers for the MCR$^2$ objective to reach a stable value. Remarkably, under the considered experimental settings, LA-ReduNet requires only approximately $1/29$ of the parameter storage of the unfolded ReduNet module for the MCR$^2$ objective to reach a stable value.

Pseudo-label-based semi-supervised learning has achieved strong performance due to its simplicity and scalability. However, it is typically developed under a closed-world assumption that unlabeled data are drawn from the same distribution as labeled data. In practical deployment, unlabeled data are often collected from open environments and may contain OOD samples. Under such contamination, OOD samples may still receive high-confidence predictions and be incorporated into training as if they were valid target examples. This creates an important evaluation problem: clean in-distribution test accuracy may appear stable even when the internal learning dynamics of SSL have already deteriorated. To address this issue, we study hidden collapse in pseudo-label-based SSL under open-world unlabeled contamination from a diagnostic evaluation perspective. We present C-Score, a compact framework that evaluates training behavior in three complementary spaces: prediction, feature representation, and optimization. C-Score includes PLE and CCI for unlabeled prediction behavior, Sem-Drift for deviation from labeled semantic anchors, and Grad-Align for the compatibility between labeled and unlabeled optimization. Experiments on CIFAR-10 and CIFAR-100 with multiple OOD sources, varying contamination ratios, and four pseudo-label-based SSL algorithms show that C-Score metrics reveal hidden degradation that clean accuracy alone fails to detect: under SVHN contamination, CCI rises over 280% while best-accuracy remains within 3% of the uncontaminated baseline; near-OOD sources (CIFAR-100, STL-10) cause up to 14.9% accuracy collapse (FlexMatch, r=0.5). The results suggest that clean accuracy alone is insufficient for evaluating SSL robustness in open-world environments, and that internal diagnostic signals are necessary for more reliable robustness assessment under unlabeled contamination.

While Digital sky surveys provide excellent throughput of image data and can cover a large footprint, their imaging power is normally inferior to that of space-based telescopes. Space-based telescopes, on the other hand, provide excellent imaging power and can image the deep Universe, but cannot provide the same throughput as advanced ground-based sky surveys. Here, we utilize generative AI to elevate the quality of galaxy images taken by ground-based telescopes to the level of details enabled by space telescopes. The solution is based on the nature of galaxy shapes, allowing generative AI trained on space-based images to convert weak signal into detailed and clear galaxy images. The method allows for combining the high throughput of ground-based sky surveys with the image quality of space-based telescopes. The source code for the method is available, as well as paired training data and a catalog of 63,202 galaxy images enhanced by the proposed method. We also provide a software tool that encapsulates the entire pipeline and the custom generative AI model to generate galaxy images with enhanced quality.

DreamBench-SWE is a multi-session benchmark for software-agent memory hygiene in which later software tasks depend on non-inferable evidence from earlier sessions and are scored by executable hidden oracles. We report the original scaled v2 fold and a separately preregistered v2.1 successor audit designed after that study but frozen before successor outcome inspection. The successor run completed 360/360 work units and 720/720 S3 cells across four conditions. In the original fold, the primary DF-hybrid–B5 contrast was null (95/180 versus 89/180; clustered p=.518, Holm p=1), not evidence of equivalence, and C9/C10 retained B0-headroom limitations. In the successor, no external memory achieved 21/180 passes (rate 0.1167), deterministic verbatim event memory 82/180 (rate 0.4556), the typed-plus-raw reference probe 83/180 (rate 0.4611), and one pinned hosted Mem0 literal-storage configuration 97/180 (rate 0.5389). The registered six-slot Family A retained unavailable slots at p=1; all three available comparisons against no memory rejected after Holm correction. Both preregistered mechanism contrasts were unavailable after pre-evaluation conformance rejection. The secondary literal-storage-versus-verbatim comparison was nonconfirmatory and sensitivity-dependent, while the comparison with the reference probe did not reject. The audit therefore supports DreamBench-SWE as a discriminating executable profile benchmark and characterizes one exact hosted-memory configuration, but it does not establish an external-system mechanism, superiority among memory-bearing conditions, equivalence, or broad product generality. The original v2.0.5 findings and artifacts remain unchanged.

Public datasets for agricultural disease detection are usually judged fit for use from reported metrics, which say nothing about whether the annotation scheme is internally consistent. On one public grape disease dataset (3288 images, 11995 boxes, 6 classes), varying model capacity, input resolution and detection paradigm yields a test-set mAP50 range comparable to seed-to-seed noise, with the bottleneck at small objects across all five architectures. The finding lies on the data side: one class is annotated at whole-leaf level (median box area 43.16% of the image) while the other five are annotated at lesion level. On 5156 cross-species images containing no grape, 65.7% of the false-positive boxes fall into that one class, an over-representation of 13.41x relative to its share of the training annotations. Counterfactual retraining establishes a causal effect of granularity on the magnitude of the shortcut: shrinking only that class's boxes cuts its cross-species false positives by 66%, and a placebo control confirms the effect is specific to the manipulated class. A manipulation in the opposite direction, with criteria registered in advance, returns a negative result: coarsening the finest class to whole-leaf level (0.57% to 40.37%), matched in box count and share of annotations and with higher in-distribution AP, still leaves its cross-species false positives at zero boxes, while the unmanipulated original class holds 50.0% of them. Annotation granularity is therefore a modulator of this shortcut, not its cause: it can amplify or attenuate a sink that already exists, but cannot create one, and what fixes the destination remains open. We also give a granularity screening statistic requiring neither images nor training, and show airborne lesion-level detection to be optically out of reach. The failure mode is invisible to in-distribution evaluation.

Surface hopping is an attractive mobility mode for small-body exploration, but designing fuel-optimal hops on asteroid 433~Eros requires solving a nonconvex optimal control problem with an irregular polyhedral gravity field, thrust–mass coupling, and collision-avoidance constraints. We show that a physics-informed Bayesian Optimization (BO) warm-start—a Gaussian-process search over a single Bézier control point that requires no offline training—provides a more reliable initialization for Sequential Convex Programming (SCP) than the standard straight-line guess. The straight chord penetrates the asteroid on every inter-site transfer considered here and degrades convergence. The physics-informed BO reference feeds an SCP stage in which a log-mass change of variables convexifies the thrust–mass coupling and nearest-facet half-spaces enforce collision avoidance, and an automated Pareto time-of-flight sweep selects fuel-priority solutions without introducing bilinear terms. Applied to all 20 ordered transfers among five representative surface sites and validated under 10 random seeds, the framework tracks every trajectory to meter-level terminal accuracy in closed-loop Monte Carlo simulation, completes a five-site tour for one fifth of the propellant budget, and reduces mean $\Delta V$ by roughly one third relative to an idealized two-impulse ballistic trajectory estimate. A straight-line ablation credits the warm-start with cutting the mean SCP iteration count from 8.8 to 5.9 and removing the one convergence failure. A target-perturbation analysis further shows that the solutions vary smoothly with the landing target, with no jumps between local basins.

Enterprise adoption of large language models in finance is constrained less by fluency than by trust: in Financial Planning and Analysis (FP&A) and other regulated workflows, an answer is usable only if it is traceable to authoritative sources and auditable after the fact. This paper argues that retrieval-augmented generation for enterprise finance should be evaluated on auditability alongside accuracy, and presents the Knowledge-Driven Analytics Framework (KDAF), which builds ontology-driven knowledge systems through six iterative stages and retrieves evidence via Context-Aware Relevance Propagation (CARP), so that every retrieved fact carries its relationship type, confidence, and source lineage. An evaluation on FinanceBench (145 questions) compares KDAF against zero-context inference, BM25, concept-weighted lexical retrieval, and ungrounded graph traversal. First, retrieval is necessary: zero-context inference reaches 4.1% correctness against 10-12% for retrieval-augmented conditions. Second, on answer correctness the retrieval conditions are statistically indistinguishable (KDAF vs BM25: -0.007, 95% CI [-0.021, 0.000]), so accuracy alone does not justify structured retrieval here – a negative result we report explicitly. Third, on auditability the ordering reverses: KDAF attains the highest citation traceability F1 (0.515), exceeding ungrounded traversal by +0.027 (CI [0.006, 0.050]) and BM25 by +0.052 (CI [0.024, 0.083]), intervals excluding zero. Graph-structured retrieval also admits no evidence from outside the question subject entity (0 of 426 items, against 16.8% and 20.2% for lexical baselines), and every selected item resolves to a complete provenance chain. We argue that auditability, not accuracy, is the axis on which ontology-grounded retrieval earns its cost.

Simulating a continuous-time quantum walk (CTQW) on a graph in the circuit model of quantum computing requires decomposing its Hamiltonian into terms that can be Trotterized into hardware-native gates. We consider two such decompositions: the standard Pauli decomposition and the recently introduced matching decomposition. Prior work suggests that the matching decomposition uses fewer CX gates on sparse graphs, while the Pauli decomposition uses fewer on denser graphs. Since CX gates dominate error and runtime on current hardware, we train machine learning models to predict, for a given graph, which of the two decompositions produces the smaller CX gate count. We train and evaluate on the complete population of all 11,117 connected eight-vertex graphs from Brendan McKay's database, so the class balance and overlap are measured directly rather than estimated. We use twelve features: ten topological properties of the graph and two that count the terms the Pauli and matching decompositions produce (n_Pauli and n_match), both computable without transpiling the simulation circuit. Standard topological properties alone provide little predictive power. Instead, the dominant signal comes from n_Pauli, a property of the Hamiltonian decomposition rather than an intrinsic property of the graph; degree variance is the only other feature that carries signal. Across a range of models the Matthews correlation coefficient (MCC) falls in a narrow band, from 0.569 untuned to 0.593 after tuning, so no single architecture stands out. We adopt a single-hidden-layer neural network at MCC 0.593. Applied frozen to a held-out, class-balanced test set of larger graphs (up to 256 vertices) from structured and Erdos-Renyi families, the model transfers, with MCC rising from 0.785 at N=8 to 1 at N>=64.

Methods that transfer predictions from two-dimensional foundation models into three-dimensional segmentation are commonly grouped by task or representation. Those groupings obscure the decisions that determine whether a system remains coherent across views: where image evidence is grounded, when observations become one identity, how semantic and granularity conflicts are handled, which information is fused, and what state survives for later queries. We introduce Lift, Associate, and Fuse (LAF), a decision-centric framework that represents a transfer system as five operators: Generate, Associate, Reconcile, Fuse, and Persist/Query. LAF defines an explicit contract for the persistent carrier—its spatial support, semantic state, identity state, uncertainty, provenance, and supported operations—and identifies the first stage at which discarded evidence becomes unrecoverable. We operationalize the framework as a structured audit protocol and apply it to 161 systems available through 7 August 2026, spanning point-, field-, Gaussian-, object-, graph-, and memory-based carriers. Representation, temporal, relational, and feed-forward stress tests required no additional analytical stage after the final confirmation pass. The resulting decision traces expose four recurring properties: association does not establish identity; carrier design fixes both the query interface and correction boundary; rendered-view, native-3D, and proposal-level evaluations are not interchangeable; and qualifiers such as training-free, real-time, open-vocabulary, and generalizable are meaningful only when attached to a stage and a complete cost ledger. LAF therefore supplies a representation-neutral method for comparing existing systems, diagnosing irreversible failures, and specifying revisable 3D perception for future agents.

LLM agents routinely construct tool-call arguments from user profiles, conversation history, retrieved documents, and prior tool results. However, legitimate access to contextual information does not imply authorization to transmit that information for every purpose or destination. We present Claw in Plain Sight, an authority- pressure attack in which task-adjacent content frames protected attributes as operationally or procedurally required, causing a model to include them in otherwise valid generated arguments. We evaluate Claw in Plain Sight using a controlled synthetic benchmark that crosses six pressure levels with four privacy-policy levels across five DeepSeek and Claude model configurations, producing 120 calls. Across the complete pressure-policy matrix, session-level disclosure rates range from 20.8% to 75.0% among the tested models. Stronger privacy instructions reduce aggregate disclosure but do not eliminate it consistently across models, showing that prompt-level policies do not provide a portable enforcement boundary. Our experiments use only synthetic profiles and capture proposed arguments locally; they measure policy-violating generation at the context-to-argument boundary, not completed network exfiltration or leakage from deployed users. These findings motivate purpose- and destination-aware inspection of generated tool arguments before execution.

Traffic sensors commonly record flow, speed, and occupancy, but standard traffic flow forecasting benchmarks and models rarely exploit all three raw measurements reliably. Although speed and occupancy provide sensor-native traffic-state information beyond flow alone, existing releases often omit these variables, replace them with proxies, or contain logically inconsistent records. Moreover, direct empirical risk minimization over three-variable inputs may exploit regime-dependent shortcuts, as the relationships among flow, speed, and occupancy vary substantially between free-flow and congested states. We introduce PEMSB-3V, a public benchmark suite that preserves raw flow, speed, and occupancy measurements from PeMS detectors for flow prediction. We also propose RiskTraf, a model-agnostic risk-extrapolated residual plug-in. For each trained spatio-temporal backbone, RiskTraf freezes the selected checkpoint and learns a lightweight zero-start residual head from historical speed and occupancy. The residual head constructs ordered traffic-risk environments and optimizes horizon-wise flow corrections with a risk extrapolation objective, thereby mitigating regime-specific shortcut correlations without modifying the backbone. Extensive experiments demonstrate that RiskTraf consistently improves diverse forecasting backbones and outperforms debiasing and distribution-shift adaptation methods. Our code and benchmark are available at https://github.com/Guangyu4/RiskTraf.

Planning and control of fixed-wing Unmanned Aerial Vehicles (UAVs) are challenging due to nonlinear dynamics, aerodynamic limits, and environmental disturbances. Differential flatness offers a principled way to generate fast, feasible trajectories, but its use has largely been confined to model-free controllers, which lack predictive capabilities and demand tuning. In this paper, we propose a unified framework that integrates differential flatness-based trajectory generation with Nonlinear Model Predictive Control (NMPC), combining computationally efficient planning with predictive, constraint-aware control. To further improve robustness, we introduce a wind-aware sampling strategy embedded within the NMPC framework, enabling the generation of dynamically feasible reference trajectories that proactively account for wind disturbances while strictly enforcing aerodynamic and control input constraints. We validate the proposed framework through extensive simulations and real-world flight experiments, demonstrating improved tracking accuracy and robustness for complex trajectories, particularly when using the proposed wind-aware sampling strategy under strong wind conditions.

Clustering methods have been used to identify distinct groups of milk samples, cows, or herds. Fourier-transform infrared (FTIR) spectroscopy, particularly mid-infrared (MIR) spectroscopy, has been applied to individual cow milk samples to predict various milk traits. Applying clustering directly to MIR spectral data may reveal latent groups of cows associated with milk traits or health disorders and can help prevent these conditions or monitor at-risk animals. This study aimed to identify groups of individual dairy cows in early lactation directly from milk MIR spectra and to analyze their associations with milk traits. Using a dataset of 407,632 individual milk MIR records from 3,408 commercial farms, we combined (i) spectral filtering that selects informative wavenumbers, (ii) two dimensionality-reduction methods: principal component analysis (PCA) and an autoencoder, and (iii) two clustering algorithms: k-means and spectral clustering to yield eight different clustering approaches. We regrouped the assigned clusters into meta-clusters that encompassed the most similar ones identified by the eight approaches. Our results revealed five distinct meta-clusters of early-lactation individual dairy cows significantly associated with milk traits. Despite substantial differences, the eight approaches converged on the same five meta-clusters, and the classic, computationally efficient PCA-based k-means approach using the full spectrum recaptured clusters identified by more sophisticated, computationally intensive approaches. The five meta-clusters were strongly associated with DIM and appeared to reflect a gradient of negative energy balance (NEB) severity: severe, moderate, and possibly mild, while the remaining two likely represented cows recovering from NEB, one with rapid restoration of energy balance and one in early recovery.

Low Earth orbit (LEO) mega-constellations create a highly non-stationary interference environment that cannot be accurately captured by static stochastic-geometry snapshots. This paper proposes a kinetic interference field framework that models the constellation as a compressible fluid shell evolving under orbital kinematics. By mapping satellite motion into a continuum flux field, we derive a hydrodynamic conservation law for the aggregate interference and obtain a closed-form expression for the time-varying outage probability via moment matching. The analysis reveals that high-latitude “interference surges” are a direct consequence of orbital compression and boundary flux, rather than random anomalies. Numerical validation against ephemeris-driven Monte Carlo simulations confirms the accuracy of the framework across time evolution, latitude, and design parameters. Leveraging the closed-form model, we further show that the conventional $90^{\circ}$ polar-orbit design is not universally outage-optimal. Instead, an inclination angle near $79^{\circ}$ at low altitude achieves a favorable trade-off between coverage continuity and geometric interference isolation. The proposed framework provides a tractable analytical tool for interference-aware 6G non-terrestrial network (NTN) design.

Designers and policymakers in sociotechnical domains like content moderation, privacy interfaces, recommender systems and beyond, must choose among a growing menu of proposed interventions, but typically lack a principled basis for comparing them. Prior work tends to evaluate interventions individually and mostly along the effectiveness criteria, while implementation constraints such as cost, effort and feasibility are often considered separately. We present a multi-criteria framework for evaluating sociotechnical interventions. This framework is instantiated through the case of misinformation, a domain of intense focus for proposed countermeasures. We survey $N=39$ researchers on 40 operationalized interventions across five evaluative criteria: political feasibility, effectiveness, user acceptance, cost, and implementation effort. We find that the interventions that experts judge to be the most effective are not always the most acceptable to the public or the most feasible to implement. We also discuss how this tension has implications for the design of sociotechnical interventions beyond misinformation, and offer a decision framework for practitioners navigating the trade-offs of sociotechnical interventions.

Splitting a bidirectional LSTM's contextual representation into a forward-only $F_i$ (strictly a function of tokens $1..i$) and a backward-only $B_i$ (strictly a function of tokens $i..n$) beats either alone and beats a fused self-attention representation for dependency relation-type classification. But a specific, natural extension of this idea – pairing a token's forward state against a candidate's backward state (“cross-direction” pairing, $F_i$ vs.\ $B_j$) – consistently underperforms same-direction pairing, and the penalty grows, not shrinks, with token distance, both paired-bootstrap significant. We diagnose why using a frozen-trunk methodology: architectural information leakage between directions is impossible by construction (a single-layer BiLSTM, verified by code inspection); 93% of the same-vs-cross gap survives freezing the trunk and training only fresh heads, ruling out training-co-adaptation as the primary cause; linear regression shows partial representational redundancy between $F_i$ and $B_i$ ($R^2{=}0.324$ vs.\ $0.028$ for a shuffled control) and a linear probe shows partial anticipatory encoding of upcoming tokens in $F_i$ (36.5% vs.\ 17.2% majority baseline) – real effects, but neither alone, nor combined, cleanly explains the full gap. Extended frozen-trunk diagnostics (a positional probe and a distance-decay probe) show directional information is genuinely stored but not exactly positioned, and propagates only a few tokens before decaying to baseline – consistent with, and mechanistically underneath, the distance-growth finding.

Operating systems (OS) serve as the foundation for all other software systems, and thus defects in OSes can lead to severe conquences, such as system crashes and data corruption, affecting billions of users. This broad impact underscores the necessity and importance of ensuring OS quality. Understanding the characteristics of OS defects is a fundamental step in this quality assurance task, as it facilitates the design of effective defect detection and debugging approaches. In this work, we conduct a large-scale study of 1,500 defects from three distinct and representative operating systems (Android, Linux, and HarmonyOS) spanning both mobile and desktop environments. To the best of our knowledge, this is the largest study of its kind in this domain. By analyzing the distribution of OS defects across multiple classification dimensions, including the OS layer where defects occur, the functions they affect, how they are triggered, their severity, and the code elements involved in their repair, as well as performing joint analysis between dimensions and cross-OS similarity comparisons, we summarize several major findings that contribute to a comprehensive understanding of OS defects across systems. Based on these findings, we provide a series of actionable implications for better OS defect detection and debugging, offering guidelines for future research aimed at improving OS quality assurance.

Multi-merchant e-commerce catalogs contain equivalent and related products under different merchant-scoped identifiers, fragmenting behavioral evidence across merchants. Expert-defined taxonomies, meanwhile, are often too coarse for fine-grained discovery. We investigate whether a single hierarchical Semantic ID (\sid{}) representation can support personalized ranking and query reformulation. Learned once from product-content embeddings, the hierarchy defines product concepts at multiple granularities that each application combines with its own behavioral and serving context. For ranking, we aggregate consumer affinity and product performance over \sid{} prefixes and derive sequence features for candidate products and consumer histories. Controlled ablations show improved offline relevance, while online evaluation of the full ranking treatment shows stronger top-slot add-to-cart engagement and broader exposure for less-popular products. For query reformulation, we ground queries and session transitions in \sid{} concepts, use the hierarchy for navigation and refinement, and filter suggestions against the merchant's assortment. Offline evaluation shows finer intent preservation than taxonomy and higher-quality suggestions than raw query-string transitions; online evaluation shows reduced search effort and earlier access to purchasable products. These results show that a shared semantic product hierarchy can support both recommendation and search while preserving the task-specific context required by each application.

Multi-View Pedestrian Detection (MVPD) aims to detect pedestrians in the form of a bird's eye view map from multi-view images. Recent MVPD methods adopt a unified framework that projects 2D image features into a 3D world space and aggregates them into a single feature. Although they are effective, they struggle to generalize to unseen camera configurations during training due to two main issues. First, they are difficult to capture accurate visual geometry across views in unseen camera configurations. Second, they make detection models highly dependent on distortion patterns during training arising from their image feature projection. To address these, we leverage a visual geometric foundation model and propose MV2GF. This foundation model has exhibited strong generalization in capturing visual geometry across views and predicting accurate 3D attributes in diverse camera configurations. MV2GF fuses task-specific features with general-purpose geometric features extracted by the foundation model to effectively capture the visual geometry even in unseen camera configurations. Furthermore, MV2GF projects each pixel in the image features to an appropriate 3D location using 3D pointmaps predicted by the foundation model, preventing the detection model from depending on distortion patterns during training. Our experiments demonstrate the effectiveness of leveraging a visual geometric foundation model for MVPD and that MV2GF generalizes better than existing methods.

The edge-of-stability (EoS) phenomenon of full-batch Adam has been widely observed, yet its underlying dynamical mechanism remains poorly understood. In this paper, we identify Adam's second-moment adaptation as a negative-feedback mechanism that drives the dynamics toward the stability boundary. We characterize this mechanism through the *active curvature*, namely, the preconditioned curvature along the preconditioned gradient direction, and establish rigorous characterizations in progressively richer settings: rank-one quadratics with momentum, diagonal quadratics, on which the active curvature separates from the sharpness, and general objectives. Importantly, the mechanism predicts *gradient reversal* of full-batch Adam near the edge: consecutive gradients repeatedly point in nearly opposite directions, as we observe across fully connected networks, ResNets, ViTs, LSTMs, GPT-2 medium, and Adam-family optimizers. Consistent with this picture, averaging iterates suppresses these fast oscillations and produces smoother and lower loss curves. Together, these results provide an important first step towards fully understanding the dynamical behavior of Adam's EoS through active curvature and gradient reversal.

Static analyzers have been widely adopted for vulnerability detection in C/C++ programs. Query-based static analyzers (e.g., CodeQL) encode vulnerable code patterns in detection queries and match them against source code. However, existing queries still suffer from false positives (FPs, incorrectly flagging benign code as vulnerable) and false negatives (FNs, missing real vulnerabilities). We present ARQ, an agentic framework that automatically refines C/C++ CodeQL queries using execution-grounded evidence from synthesized C/C++ programs. Our key insight is that a synthesized program exposes a query's weakness whenever its execution disagrees with the query's verdict. If the program is genuinely vulnerable but the query stays silent, the query has an FN weakness; if the program is safe but the query fires anyway, it has an FP weakness. ARQ then runs an LLM-based refinement loop that repairs the query using these disagreements as ground truth. Unlike previous query refining methods, ARQ requires no labeled datasets, no commit history, and no vulnerability-specific templates. We demonstrate the effectiveness of ARQ by refining 12 official CodeQL queries using three commercial LLMs (GPT-5.4, Claude-Sonnet-4.6, and Gemini-3.5-flash). We compare both ARQ-refined and original CodeQL queries on the Juliet v1.3 and FormAI v2 datasets and show that ARQ-refined queries detect substantially more true positives, by up to 119.8%, with a Precision of at least 98.0% throughout. ARQ successfully fixed three unresolved GitHub issues raised in the official CodeQL query repository that had remained open for as long as 27 months. The refined queries also exposed two previously undiscovered bugs in the real-world libraries libpng and zlib.

Many text classification decisions are viable based on constituent excerpts alone. Taking inspiration from the field of multiple instance learning, we present an algorithm for training a neural network to classify text by selecting such excerpts. We show that our approach is also scalable with demonstrated learning against samples with nearly 1M tokens. We evaluate our methods on 7 datasets with emphasis on long-textual collections that far exceed the encoding limit of our base model. We present state-of-the-art results with this algorithm on 3 datasets: identification of political bias in news outlets, trigger warnings in long stories, and demographic characteristics of authors in tweet collections. Furthermore, the model trained on weakly-labeled collections of text (bags) generalizes to accurately classify constituent, smaller instances. Besides a new state-of-the-art for these problems, this approach is one of the few neural methods to excel in these datasets.

Discrete diffusion models have demonstrated strong performance across a range of datasets, including natural language data and graph-structured data. Among many variants, score-entropy discrete diffusion (SEDD) has achieved particularly strong empirical results. In SEDD, new samples are generated by iteratively evaluating a sequence of concrete score functions, which are learned by minimizing a score-entropy loss. While much of the prior theoretical literature on discrete diffusion has focused on the sampling efficiency of SEDD under the assumption of small score estimation error, recent work has begun to investigate the finite-sample properties of score estimation itself. In this work, we take a different route by investigating the fundamental statistical limits of concrete score estimation. We focus on uniform and masking discrete diffusions, two of the most widely adopted discrete diffusion models. We establish a minimax lower bound under the score-entropy loss, and propose an MLE-based thresholding estimator that matches this lower bound up to constant and polylogarithmic factors that depend on neighboring density ratios. We further show that, for any target distribution, this density ratio is naturally controlled under both uniform and masking discrete diffusion models, yielding nearly matching minimax lower and upper bounds for the aggregated score estimation error. Our results imply that, with appropriate initialization and discretization, SEDD can achieve nearly optimal minimax sample complexity, as measured by the KL divergence between the target and generated distributions.

Agents learn to act through interaction with environments, yet the environments used for training are often manually constructed or synthesized around predefined tasks and benchmarks. This task-centric paradigm makes it difficult to scale environments that reflect realistic and evolving workflows where diverse tasks can naturally emerge from the underlying world. We introduce AgentMercury, a scalable framework for synthesizing executable environments from high-level business scenarios. Rather than constructing an environment for a specific task, AgentMercury first instantiates a persistent world with entities, services, tools, state, and executable cross-service invariants, from which diverse tasks and interaction trajectories can subsequently emerge. We construct 4,783 executable environments spanning 14 industries and 50 countries, and use them as training substrates for reinforcement learning. Despite being generated without targeting the evaluation benchmarks, policies trained on these business-oriented environments improve substantially on both enterprise workflows and out-of-domain benchmarks spanning reasoning, coding, scientific computing, and tool use. In our experiments, Qwen3.5-4B improves from 12.3 to 15.7 on EnterpriseOps-GYM and from 45.9 to 56.0 on AIME26 after training on AgentMercury environments. We further show that the construction process itself can be learned: fine-tuning Qwen3.5-35B-A3B on construction traces increases executable-world authoring success from 3.3% to 83.3% on held-out business scenarios. These results show that scenario-grounded environments can provide useful and generalizable learning signals beyond benchmark-specific training, while their construction can itself become a learnable capability.

The randomly pivoted Cholesky algorithm is one of the leading methods for computing a low-rank approximation to a large positive-semidefinite matrix. However, while it consistently achieves accuracy comparable to or better than competing methods of its type in experiments, its theoretical analysis lags somewhat behind other methods. This paper closes this gap, proving that randomly pivoted Cholesky produces an approximation with expected error within a $1+\varepsilon$ factor of the optimal rank-$r$ approximation in $\mathcal{O}(r/\varepsilon + r\sqrt{\log r})$ steps. This result nearly matches the optimal complexity $\Theta(r/\varepsilon)$ for any low-rank approximation method based on a partial Cholesky decomposition (also known as a column Nyström approximation). The paper also presents bounds on the randomly pivoted Cholesky trace and spectral-norm errors that hold with high probability. The mathematical argument is largely due to GPT 5.6-Sol (Pro), with some refinements by the author.

Efficient-transformer research often motivates token pruning and adaptive computation with the claim that not all tokens require equal computational effort. We test this claim end to end using SEWN, a two-stream Transformer that routes tokens through either lightweight or full-capacity processing using a learned gate. Across our experiments, routing introduces negligible accuracy change relative to parameter-matched baselines, while the gate's token-importance signal depends critically on how it is learned. A static lexicon-seeded prior fails a counterfactual faithfulness test on BoolQ, whereas a fully contextual gate achieves highly significant separation ($p<10^{-10}$) on both evaluated tasks without changing task accuracy.

Large language model (LLM) agents have demonstrated the ability to solve multi-step tasks requiring planning, tool use, and external information access, yet growing execution histories increase inference cost and expose reasoning to outdated, irrelevant, or misleading information, potentially degrading reasoning quality. Existing memory approaches organize or compress execution histories but provide limited mechanisms for deciding which memories remain active. We introduce the, a hierarchical memory system that organizes execution into tasks, subtasks, and actions while assigning each memory a dynamic retention score. Event-based updates and selection-based decay revise these scores, allowing WMT to preserve useful information, fold completed trajectories, suppress low-utility content, and retain access to folded context. We evaluate WMT on GAIA-Text using Qwen3-8B, Gemma 4 E4B, and Llama-3.1-8B, with ablations and memory-poisoning experiments. Relative to linear memory, WMT improves accuracy by an average of 9.97 percentage points while reducing prompt-token usage by 32.8%. Memory-poisoning experiments show that WMT limits the persistence and propagation of unreliable information. Our results suggest that effective long-horizon agent memory depends less on storing more information than on deciding which information should remain active.

SQL systems increasingly expose AI functions for tasks such as classification, extraction, filtering, ranking, retrieval, joining, and summarization. Despite their diverse APIs, these functions play only three relational roles: transforming individual rows, aggregating groups, or generating relationships between row pairs. We present SAGE (Self-Adaptive Generative Execution), a unified logical and physical framework that captures these roles with three typed primitives, AI_SCALAR, AI_AGG, and AI_JOIN, and composes them naturally with standard relational operators. All primitives share a confidence-gated execution interface while supporting physical strategies tailored to their relational shape. The main challenge is AI_JOIN, where SAGE analyzes the predicate, decomposes compound conditions when possible, and uses a recipe card together with a small label-free probe to select among complete execution strategies. Across a broad audit of public AI operators and evaluations spanning scalar, aggregate, and join workloads, this formulation covers common AI functionality while consistently improving execution quality and efficiency. SAGE achieves the strongest overall SemBench performance and, on a representative factorable join, reduces pairwise model calls by more than two orders of magnitude, yielding a 358-fold measured cost reduction.