Robotics
Papers
Fresh from arXiv, typeset properly. Highlighted words trade on the market.
Tuesday 11 August
Contact-based industrial inspection requires aerial platforms to maintain stable interaction while rejecting disturbances. Underactuated aerial manipulators present control challenges due to the dynamic coupling between vehicle attitude and force generation. This paper proposes a robust control framework for an underactuated hexarotor equipped with a 1-DoF manipulator to perform sustained contact inspection. The architecture integrates integral-augmented Sliding Mode Control (SMC) for trajectory tracking with an admittance control law for force regulation. The contact force is mapped to a feedforward attitude term, while the 1-DoF arm actively compensates for the tilt to maintain surface alignment. Software-in-the-loop simulations demonstrate that the SMC-based approach achieves superior tracking and coupling rejection compared to traditional PID. Furthermore, the interaction strategy achieved precise force regulation with an RMSE of 0.12N and was able to stably exert up to 20N force, confirming the system's efficacy for stable, reliable contact-based inspection.
Robotics
OAA: Three Phases of Vocal Guidance in Human-Drone Teleoperation
Voice-guided teleoperation requires systems that adapt to the evolving dynamics of human guidance. Yet most voice-controlled robot systems treat spoken commands as a stationary stream, ignoring how the guide's communicative behavior changes as the task progresses. Using motion capture and speech data from two experimental configurations, humanhuman guidance (finger pointing, N =10 dyads) and humandrone teleoperation (gamepad control, N =29 dyads), we show that spontaneous vocal guidance consistently organizes into three kinematically and linguistically distinct phases: Orientation, Approach, and Adjustment. These phases are identified automatically via change point detection on 3D trajectory signals, and validated statistically (Kruskal-Wallis, p<.001). Three lexical families replicate across configurations: rotation vocabulary marks Orientation, translation vocabulary is scarce there, and attenuators accumulate toward Adjustment. Together with inter-utterance silence, these cues mark the Orientation boundary that speech rate alone leaves unmarked. The same three-phase structure emerges in both configurations despite radically different motor interfaces, suggesting it is an intrinsic property of human spatial guidance rather than an artifact of the experimental setup. We discuss implications for OAA-aware adaptive control in voice-guided teleoperation.
Read the full abstract
Artificial Intelligence
Decision-Aware Approximation of Belief Functions for Evidential Combinatorial Optimization
Reducing the number of focal elements of a mass function is classically driven by an intrinsic distance, such as Jaccard or Jousselme, that keeps the approximation close to the original as a body of evidence. We consider instead the case where the mass function feeds a linear combinatorial optimisation problem with evidential costs. What should then be preserved is not the closeness of the two mass functions, but the quality of the decision they induce. We introduce a decision-aware approximation that targets the regret of the decision: one decides with the cheaper approximation and is evaluated under the true mass function. On a minimal shortest path, the distance-optimal approximation flips the decision while a decision-aware merge preserves it, and this occurs on a non-negligible fraction of random instances. We prove a one-point bound that localises the regret at the true optimum, turn it into an exact dynamic program for the scalar case, and extend it to an online version that prunes focal elements before the final cost is known. In experiments the decision-aware compressor flips the decision less often than representation-aware compression, for both the linear criterion and a non-linear proxy read-out.
Read the full abstract
Computer Vision
PolypVision: A Three-Stage Hierarchical Deep Learning Framework for Classification and Segmentation of Colorectal Polyps
Colorectal cancer (CRC) remains one of the leading causes of cancer-related mortality worldwide, predominantly arising from precancerous polyps. Accurate detection, segmentation, and endoscopic and histological classification of colorectal polyps are crucial for timely clinical intervention. In this study, we present PolypVision, a three-stage hierarchical deep learning framework that sequentially performs: (Stage 1) binary classification of polyps as adenomatous or hyperplastic, with simultaneous Paris and JNet classification, using EfficientNetV2-M with Focal Loss; (Stage 2) polyp segmentation with recommended resection method using a UNet++ decoder with the Stage 1 backbone as encoder, optimized with Dice and BCE losses; and (Stage 3) adenoma subtype classification (tubular, tubulovillous, villous) using EfficientNetV2-M with transfer learning from Stage 2. Evaluated on three public datasets – PolypGen, Kvasir-SEG, and CVC-ClinicDB – PolypVision achieves an AUC of approximately 0.99 for frame classification and a detection mAP@50 of 94.4% on Kvasir-SEG, outperforming or matching state-of-the-art methods. Gradient-weighted Class Activation Maps (Grad-CAM) confirm that the model attends to clinically relevant lesion features. The framework is device-independent, operating across diverse endoscopic imaging systems without hardware-specific adaptation. These results demonstrate that a hierarchical, transfer-learning-driven pipeline with task-specific loss functions offers a robust, device-independent, and clinically meaningful approach to automated colorectal polyp analysis. PolypVision is freely available as a web application at https://polypvision.com, a DataBioX initiative, with a free usage tier open to all users.
Read the full abstract
Computer Vision
Precise Top-Layer Fabric Segmentation for Fabric Destacking with Edge- and Shape-Aware Deep Networks
Fabric destacking requires precise segmentation of the topmost fabric layer, a task complicated by subtle fabric boundaries and high visual similarity between fabric layers. Existing semantic and edge-based segmentation approaches often struggle with these complexities, limiting the performance of robotic manipulation for different tasks. In this work, a novel segmentation training architecture tailored for top-layer fabric segmentation in stacked fabrics is proposed. The method extends the classical encoder-decoder framework by introducing two specialized branches - an edge-aware branch and a shape-aware branch - that are used to supervise the backbone network for better tuning. The edge-aware branch enhances boundary delineation, while the shape-aware branch guides the network to capture and align the overall fabric shape with reference masks derived from Computer Aided Design (CAD) models. Experiments on a real-world fabric dataset demonstrate that the training approach outperforms established baselines, verifying the effectiveness of the multi-branch design through both quantitative results and ablation studies.
Read the full abstract
Human-Computer Interaction
ProtoGIB-Workload: Learning Workload-Specific Neural Topology Prototypes across Subjects
Reliable electroencephalography (EEG)-based mental workload recognition is crucial for adaptive human-centered systems, yet practical deployment requires models to generalize to users unseen during training. Although functional connectivity graphs are widely adopted to capture workload-related neural interactions, they inherently entangle task-relevant structures with subject-specific physiological traits and sample-level noise. This entanglement often leads models to learn structural shortcuts, severely degrading cross-subject generalization. To address this, we propose ProtoGIB-Workload, a novel framework that explicitly regularizes and aligns graph structures for subject-independent workload recognition. Our approach introduces a Stochastic Graph Information Bottleneck (SGIB) to compress dense correlation priors into compact, task-relevant subgraphs, filtering out input-related redundancy. Crucially, to prevent the retention of subject-specific spurious edges, we propose a Class-Conditional Topology Stabilizer (CTS). Leveraging the fixed electrode coordinates of EEG data, CTS operates directly on graph-generation probabilities to encourage consistent edge-generation statistics across different subjects sharing the same workload class. Extensive experiments on two public EEG workload datasets and one in-house EEG cognitive load dataset of air traffic controllers under strict leave-one-subject-out (LOSO) protocols demonstrate that ProtoGIB-Workload significantly outperforms state-of-the-art temporal and graph-based baselines, improving the cross-subject Macro-F1 score by an average of 5.15% (up to 6.34%). Further analyses confirm that our method successfully extracts stable, cross-subject consistent neural connectivity patterns.
Read the full abstract
Multiagent Systems
ASCon: A Direction-Aware Reciprocal Agent–Step Contextualization Model for Failure Attribution in Multi-Agent Systems
Failure attribution in LLM-based multi-agent systems (MAS) aims to answer who caused failures, when they occurred, and why by identifying responsible targets including faulty agents, erroneous steps, and failure modes. Existing methods have primarily focused on developing dedicated models for specific attribution targets, with limited attention to the evidential dependencies among them. Despite these attribution targets are different, they rely on common diagnostic evidence from MAS trajectories, including task constraints, agent roles, behavioral histories and inter-agent interactions. This commonality motivates us to develop a unified representation model that aggregates the trajectory evidence into individual agent and step representations, which can subsequently be adapted to different attribution targets. Accordingly, we propose ASCon, a direction-aware reciprocal Agent–Step Contextualization model for multiple failure attribution targets. ASCon introduces direction-aware graph attention to model execution context, masked step-to-agent attention to construct behavior-aware agent representations, and agent-conditioned step contextualization to incorporate agent context back into step representations. The resulting contextualized representations enable different attribution targets through lightweight target-specific heads. Experiments show that ASCon can improve faulty-agent detection by 5.83%+ in micro-accuracy, faulty-step detection by 10.63%+ in micro-accuracy, and failure-mode detection by 14.73%+ in Macro-F1. Meanwhile, it can also substantially enhance the LLM-based methods' attribution capabilities in out-of-domain scenarios.
Read the full abstract
Systems and Control
AIDC Microgrid Vulnerability Assessment Under Computing-Power Coordinated Attacks
The rapid growth of large language model (LLM) services is expanding AI data centers (AIDCs), increasing electricity demand and associated carbon emissions. Renewable energy integration can mitigate these impacts but also strengthens the coupling between AIDC loads and inverter-interfaced generation, creating cross-domain cyber-physical vulnerabilities. Specifically, adversarial AI requests alter AIDC power demand, whereas inverter control tampering modifies source-side dynamics, and their combined impact on system stability varies with generation forecast and demand response uncertainties. To this end, we propose an uncertainty-aware AIDC microgrid vulnerability assessment framework under computing-power coordinated attacks. First, the framework maps adversarial AI requests to AIDC power variations and represents uncertainties in attack-induced demand responses and photovoltaic (PV) forecasts through confidence-weighted realizations. Then, impedance based stability analysis combines these realizations with bounded inverter parameter tampering to construct attack reachable domains and identify critical attack time windows. Furthermore, a separate criterion identifies fixed coordinated attack vectors that retain destabilizing capability throughout each selected window. Case studies demonstrate that, unlike either attack component applied alone, coordinated attacks within identified critical windows induce sustained inverter frequency oscillations with peak absolute deviations exceeding 20% of nominal frequency, whereas the evaluated out-of-window response remains bounded. The proposed method further identifies critical attack windows and the associated coordinated attack vectors.
Read the full abstract
Artificial Intelligence
Curate Before You Connect: Identity and Ontology Tagging in a Production Knowledge Graph
Extraction produces candidate entities and relationships; writing them into a graph is where identity is decided, and identity decisions are destructive in a way extraction errors are not. A wrong type can be corrected later, but two records merged under one identity cannot be separated once their properties have been combined, and the merge leaves no error behind. This paper describes the ingestion and ontology-tagging layer that turns a validated extraction stream into a knowledge graph of 537,157 entities and 2,198,567 relationships drawn from 98,795 government documents. We describe a record-identity ladder that decides sameness from identifier columns, name columns, display names and type-scoped position rather than from name similarity. The ladder governs de-duplication within parsed tables, while the graph write applies a coarser canonical-name key, so records sharing a canonical name merge automatically on exact equality. We argue rather than demonstrate that this is where the automation line belongs: no identity benchmark is reported, and the over-merges the key permits are undetectable by construction. That policy, under which entity resolution only ever flags candidates, followed an incident in which two surface forms of one name were merged, corrupting a correct record and deleting eight entities from an unrelated document. We then describe multi-class ontology tagging and an evidence asymmetry we did not anticipate: an entity name is an instance label rather than a type assertion, so matching name fragments against a class index invents classifications. Requiring anchored evidence cut role assignments on an enriched sample from 36 to 4, all confirmed correct. We quantify the graph's conformance debt, show secondary classifications compensating for a mis-parented primary class, and describe a curation queue grown to 48,403 pending proposals against 775 human decisions.
Read the full abstract
Systems and Control
Removal-Only Actuation in Age-Structured Branching Populations: Fundamental Limits of Equilibrium Placement
A subcritical age-structured branching population dies out almost surely. Conditioned on survival, it converges to its Yaglom limit, a quasi-stationary equilibrium that we take as the operating point for control. We model preventive removal (culling) as an age-dependent actuator that raises the mortality rate and leaves the offspring law untouched, and we show that its authority over this equilibrium is bounded for structural reasons. Two facts drive the result. First, the input is matched to the killing rate but unmatched with respect to the Foster–Lyapunov drift, so the transmission barrier $\Lbar$ set by reproduction alone is invariant under such actuation. Second, and this does not follow from invariance alone,} the supremum of the reachable decay rates is $\Lbar+\nu^\star$, where $\nu^\star\le0$ is the Malthusian parameter of the lineage conditioned never to die childless; the gap $|\nu^\star|$ is given in closed form and vanishes exactly when no individual has two or more offspring. Consequently no removal law of this class reaches the barrier, and along the admissibility boundary the achievable decay rate is governed by the shape of the actuator rather than by its size. We illustrate these results on a model calibrated to the 2001 Cumbrian foot-and-mouth outbreak.
Read the full abstract
Information Theory
Threshold-Based Spiking Neural Networks for Event-Driven Status Update Systems
Event-driven sensing supports energy-efficient Internet-of-Things (IoT) devices by activating communication only when relevant events occur. In such systems, transmission decisions are governed by the monitored process rather than predefined schedules. Consequently, jointly optimising information freshness and energy consumption is challenging because transmission decisions are restricted to randomly occurring events. To address this challenge, we investigate an event-driven status update system in which wake-up events follow the dynamics of the monitored process. The problem of determining whether to transmit the sensing data or not is cast as a Markov Decision Process (MDP) that jointly minimises the Age of Information (AoI) and transmission energy. We prove the existence of an optimal threshold policy, thereby obtaining an interpretable characterisation of the optimal transmission strategy. Motivated by this result, we propose a lightweight Reinforcement Learning (RL) approach based on Spiking Neural Networks (SNNs) whose architecture explicitly represents threshold policies. The resulting policy representation has constant complexity with respect to the maximum AoI and enables a more energy-efficient implementation than a comparable Artificial Neural Network (ANN). Numerical results demonstrate that the proposed SNN reliably learns optimal thresholds across different operating regimes.
Read the full abstract
Information Retrieval
DistilVDR: A Compact End-to-End Visual Document Retriever via Dual-Student Distillation
Visual document retrieval (VDR) is dominated by multi-billion-parameter models that are slow to index at full corpus scale and expensive to serve. Prior compression routes either train a smaller multi-vector encoder from scratch or distil only the query side; neither yields a compact single-vector retriever end-to-end. We present DistilVDR, a 524M end-to-end VDR system distilled bilaterally from a single 8B vision-language teacher under a pointwise cosine alignment loss. All supervision comes from the frozen teacher's embedding space, which was itself trained with relevance supervision, so the student objective needs no relevance labels, negative sampling, or contrastive term. We match VDR's text-query and image-document input asymmetry with an asymmetric encoder-only student that concentrates visual capacity on the document side and keeps the query side at 70M parameters. We release two variants that share the same encoders and training and differ only in the document encoder's visual-tile budget: DistilVDR-HiRes attains 61.74 average NDCG@5 on ViDoRe v1+v2+v3 (86.9% of the 8B teacher) and leads every reproduced sub-1B baseline on the high-resolution-sensitive v3 benchmark, while DistilVDR-Fast attains 59.98 at a 3 times smaller visual-token budget. Both variants store one million documents in a 15.6 times smaller index than the strongest sub-1B multi-vector baseline and index the corpus an order of magnitude faster. The code is available at https://github.com/Ryenhails/NanoVDR.
Read the full abstract
Computer Vision
MedUP: Awakening Unified Understanding and Perception in Medical Vision-Language Models
Medical Vision-Language Models (Med-VLMs) excel at verbalizing visual content, yet precise visual perception, segmentation, and grounding remain challenging. Existing approaches either verbalize regions as coordinate strings or rely on external modules that decouple perception from understanding, creating representation gaps for region-language alignment. We present MedUP, a Med-VLM that natively unifies perception and understanding within a shared token space. At its core lies UniMedTok, a region tokenizer that encodes masks as discrete tokens in the LLM vocabulary, enabling the model to seamlessly interleave mask tokens with text. We curate UniMed-Train, a 1.84M-instance corpus spanning text-guided segmentation, region-grounded understanding, medical VQA and CoT-based segmentation, and introduce UniMed-Bench for unified evaluation. Extensive experiments show that MedUP outperforms native, agentic, and dual-decoder Med-VLMs across all tasks while remaining competitive with specialist segmentors, demonstrating the strong potential of unified understanding and perception modeling.
Read the full abstract
Machine Learning
IADD-TR: Intervention-Aware Dynamics Decoupling with Targeted Regularization for Model-Based Reinforcement Learning
Model-based reinforcement learning (MBRL), which learns environment dynamics to generate synthetic experience, is a promising approach to sample-efficient decision making. Numerous methods have been developed to improve dynamics prediction and policy optimization for MBRL through uncertainty estimation, model regularization, and conservative value learning. However, these methods typically treat the transition model and critic as monolithic predictors, overlooking the policy-induced data bias. Consequently, action can become entangled with environmental evolution, while uneven action coverage may distort the counterfactual value estimates used for policy improvement. To address this, we propose IADD-TR, a unified framework combining Intervention-Aware Dynamics Decoupling (IADD) and Targeted Regularization (TR). IADD factorizes transitions into an action-intervention stage and an action-free natural evolution stage, using a zero-action anchor to resolve the non-uniqueness of this two-stage factorization for robust generalization. Its latent and state-aligned components are identifiable up to an invertible within-block transformation and pointwise, respectively. For policy learning, we derive TR from the efficient influence function of a replay-state policy-gradient functional. TR augments the critic with an action-density-scaled residual correction and optimizes a targeted loss, yielding doubly robust policy-gradient estimation when either the critic or the replay action density is consistently specified. Extensive experiments on five MuJoCo tasks show that IADD-TR achieves competitive returns with improved sample efficiency.
Read the full abstract
Computer Vision
InSight-doc: Agentic Visual Perception for Long-Document Understanding
Long-document understanding often requires reasoning over many visually rich pages, making inference costly and prone to context rot. In this work, we propose InSight-doc, an agentic visual perception framework that treats visual resolution as an adaptive reasoning-time resource. InSight-doc starts from low resolution and selectively zooms into high-resolution regions for finer evidence, without relying on any external retriever. To train such an agent, we construct an active-perception corpus of 17.9K high-quality SFT examples with region-level zoom-in trajectories, accompanied by 19.2K hard RL examples. Through SFT+RL, InSight-doc-8B improves the baseline by 4.3–16.4 accuracy points over document VQA benchmarks. On long documents, it reduces hallucination by more than 40% and inference latency by 41%–68% while maintaining an accuracy lead. Our code, datasets, and model are released at https://github.com/m-Just/InSight-doc .
Read the full abstract
Computation and Language
Decomposition-Induced Context-Memory Conflict: An Instruction-Induced Failure and a Probing Case Study
Decompose-then-verify pipelines (FActScore-style fact-checkers, hallucination detectors) first split a passage into atomic claims and treat this step as neutral preprocessing. We ask whether a decomposer can substitute its own parametric belief for what the passage says, producing a claim that contradicts its source, and call this Decomposition-Induced Context-Memory Conflict (DI-CC). On biographies with a deliberately altered year, Qwen2.5-7B-Instruct does so in 4.3% of claims when the instruction licenses self-correction, but in only 0.2-0.5% under a neutral instruction, a FActScore-style few-shot prompt, a "use only the passage" instruction, a formatting instruction, or a knowledge-filling instruction (an upper bound: in the neutral and FActScore-style runs only 1 of 13 flagged claims is a real override); context-aware decoding removes only half. DI-CC here is therefore instruction-induced. We then test whether a linear probe trained only on classical context-memory conflict data (NQ-Swap) detects DI-CC zero-shot. Its AUC is high (up to 0.94 against legit claims, 0.98 under a position- and instruction-matched control), but random directions reach AUCs between about 0.1 and 0.9 on these contrasts, probes fit directly on the claims reach 0.96-1.00 everywhere (even on a control that separates only the text variant), and after multiplicity correction the NQ-Swap probe beats random directions only at the last layer, and only for two of three random-direction families. A direct NLI check between source and claim reaches AUC 0.998. High probe AUC on such small contrasts is not evidence of a shared mechanism without nulls, and two pitfalls, pair leakage in probe training and duplicated BOS tokens, can change such conclusions.
Read the full abstract
Computation and Language
Dual-Loop Self-Evolution via Verifiable Emotion Feedback for Multi-Turn Empathetic Dialogue
Large language models have demonstrated conversational capabilities, yet empathetic competence remains challenging. Empathetic support is inherently multi-turn and path-dependent: users disclose concerns gradually, emotions evolve over time, and early responses shape trust and receptivity. Reinforcement learning with verifiable emotion rewards provides scalable supervision for long-horizon interactions. However, existing methods evolve the dialogue policy while keeping its training interaction distribution fixed, creating a mismatch between policy competence and training experience. We introduce a dual-loop self-evolution framework driven by verifiable emotion feedback. With the user simulator and verifier frozen, the inner loop optimizes the multi-turn policy using continuous emotion rewards, while the outer loop uses the same outcomes to estimate policy-relative interaction utility and adapt experience. To obtain estimates from sparse, stochastic rollouts, the framework holds the scenario and interaction state constant within each group and prioritizes conditions whose group pass rates lie near the policy's competence boundary. A hierarchical controller shares evidence across support intents, while uncertainty-guided exploration and uniform rehearsal prevent premature exclusion. The resulting distribution generates trajectories, closing both loops without increasing the rollout budget. On SAGE, our framework raises Qwen3-8B Overall from 53.87 to 79.24 and outperforms protocol-matched uniform emotion-reward reinforcement learning by 7.23 points.
Read the full abstract
Networking
User Satisfaction-Aware Resource Allocation with Prospect-Theoretic Utility for Multimedia Streaming
Traditional resource allocation policies for multimedia streaming have primarily targeted metrics such as throughput, fairness and delay. However, user satisfaction is known to be strongly influenced by variations in throughput. Although some recent works have considered throughput variations, they overlooked a well-known aspect of human behavior that has reshaped behavioral economics over the past few decades: the same amount of decrease in throughput causes greater annoyance than the satisfaction caused by the increase. Motivated by this insight from the broad field of economics, popularly known as prospect theory, we propose a generic utility (metric) that captures both the average throughput and the asymmetric effect of throughput variations on user satisfaction. Next, we propose a resource allocation policy that reduces abrupt decreases in allocated resources when the system transitions from a resource-rich to resource-scarce state while respecting the priorities of different classes of users. Extensive simulations show that this policy outperforms the existing fair allocation policies. However, the choice of the optimal threshold depends on the network instance, which, in practice, may change frequently. Hence, building on the theoretical analysis of multi-queue systems, we propose fast numerical methods for determining the optimal threshold, which can be used in field deployments.
Read the full abstract
Robotics
When Your State Estimator Has Lost The Plot: Detecting Estimator Failures Via Spectral Analysis
Reliable onboard state estimation is essential for safe robotic operation, yet unmodeled disturbances, such as sensor aliasing or out-of-distribution noise, still cause estimators to degrade or fail completely. While many methods aim to improve estimator robustness, only a few provide introspective mechanisms to assess estimate quality. Existing uncertainty measures, such as covariances, rely on idealized assumptions and tend to be overconfident, and more recent data-driven approaches are typically tied to their training data distributions. We propose a sensor-agnostic introspective method that assesses estimator health by analyzing the frequency-domain power distribution of recent velocity estimates. The method is evaluated using outdoor flight data from an aerial robot running visual-inertial, LiDAR-inertial, and radar-inertial odometry. The dataset includes multiple estimator failures, enabling analysis of several frequency-domain indicators, such as signal power, spectral bandwidth, and entropy. We observe consistent spectral power differences between healthy and degraded estimates, allowing detection of 51%-58% of labeled failures with 60%-84% precision across three fundamentally different state estimation frameworks. Our results show that even a simple frequency-domain analysis of a state estimator's output can serve as a lightweight introspective tool to complement existing robustness techniques in real-world robotic deployments, and opens promising avenues for future investigation.
Read the full abstract
Software Engineering
A Study of Cursorrules Files in GitHub Open Source Projects
Prompts are the primary mechanism for communicating with AI agents, and they directly influence the quality and reliability of AI-generated code. As AI-assisted programming becomes widely adopted, modern tools increasingly combine dynamic conversational prompts with static configuration-like prompt files. Despite the growing focus on prompt engineering, prior research has primarily focused on conversational prompts, while prompt files remain understudied. To address this gap, we conduct an empirical study of configuration prompt files in Cursor, a widely used AI-assisted code editor. We collect and analyze over 12,110 .cursorrules files from 11,427 GitHub repositories to characterize their distribution, evolution, and maintenance. Complementing this, we perform qualitative analysis on a random sample of 65 prompt files and develop a 65-code codebook capturing how developers express programming intent, project context, engineering practices, and security considerations. Our results show that .cursorrules files emerged rapidly from mid-2024. Their adoption is concentrated in small-scale, low-activity, single-maintainer repositories, suggesting toy projects rather than professional development. The content of prompt files is dominated by guidance on code quality and engineering practices, project structure and configuration, and maintainability, while security-related content appears less frequently. Our analysis shows that there is a continuity of themes and topics between the now-legacy .cursorrules files and the current standard .mdc files.
Read the full abstract
Machine Learning
ProbGuard: Calibrated Safety Risk Estimation from LLM Output Distributions
Recent research on Large Language Model (LLM) safety has widely adopted guardrails to identify unsafe LLM outputs. Existing guardrails typically formulate safety assessment as a deterministic classification task, mapping a discrete token sequence to a discrete safety label. However, this paradigm has two limitations: First, safety assessment is inherently an uncertain problem, particularly during the early generation state. Second, relying solely on discrete token sequences discards the rich probabilistic information embedded in the LLM output distribution. To address these limitations, we propose the first completely probabilistic architecture-agnostic guardrail ProbGuard to leverage the LLM early output distributional signals for estimating and calibrating the safety probability, thereby enabling early stopping of unsafe ongoing outputs. Specifically, given an LLM's generated prefix distribution, we formulate the safety risk as the unsafe probability of its continued generation dynamics and estimate this risk by Monte-Carlo sampling. Through post-training on the distributional signals and calibrated safety risk, ProbGuard achieves the best calibration performance across all nine model–dataset combination settings, reducing the average Brier score and ECE by 79.6% and 71.9%, respectively, over the best baseline. ProbGuard further limits the attack success rate to at most 1% across six representative jailbreak attacks after observing the LLM early output distributions from only the first ten decoding steps.
Read the full abstract
Machine Learning
Pair-Centric Graph Rewiring for Over-Squashing via Optimal Transport-Guided Communication Alignment
Message-passing neural networks (MPNNs) often struggle when task-relevant information is distributed across distant regions of a graph, since local propagation must compress remote signals through limited structural interfaces. Graph rewiring provides a structural response to over-squashing. Most existing methods rely on edge-level bottleneck scores or graph-level connectivity surrogates. With a limited rewiring budget, the key question is which pairwise communications most need structural support. This paper proposes PairAlign, a pair-centric graph rewiring framework that makes this question explicit through demand-support shortage. Specifically, PairAlign combines original-graph structural demand with current-graph finite-hop propagation support; their ratio highlights interactions whose communication demand is poorly supported by topology, and our theory shows that this score provides a computable proxy for the corresponding Jacobian-based shortage with a pair-level interpretation of over-squashing. Our theory reveals a two-sided effect of edge insertion: a new edge can create useful walks and simultaneously dilute existing normalized transition mass. Guided by this observation, PairAlign optimizes shortage to favor edge additions that alleviate over-squashing. Beyond selecting useful additions, PairAlign further introduces an Optimal Transport-guided rewiring mechanism to coordinate the finite edge budget for pair-level structural compatibility and shortage-target coverage. It formulates communication alignment between the candidate edge budget and the shortage targets, and the theory shows that this allocation covers shortage targets more broadly and effectively than a greedy-local assignment. Experiments on standard graph benchmarks show PairAlign's improvement across message-passing backbones, validating pair-level repair as an effective route for alleviating over-squashing.
Read the full abstract
Robotics
Toward the Cognitive–Physical Limits of Embodied Intelligence through a World-Model-Centric Autonomous Racing Agent
Embodied artificial intelligence aims to develop agents that perceive, reason, and act through continuous interaction with the physical world. However, most embodied systems are still evaluated within conservative safety margins or moderate interaction regimes, leaving their capability boundaries under extreme conditions insufficiently understood. Autonomous racing provides a stringent testbed by combining high-frequency localization and perception, adversarial interaction, near-saturated vehicle dynamics, and strict safety constraints. Existing systems push high-speed performance but rarely model and refine cognitive and physical limits jointly. Here we show that a world-model-centric autonomous racing agent provides a concrete step toward exploring these coupled limits. The framework learns predictive world models from near-limit successes and failures to capture interaction evolution, ego dynamics, and feasible-motion boundaries, coupling world-state construction, future-aware reasoning, and near-limit control in a closed-loop refinement process. Training data were collected from real-vehicle autonomous racing, where the onboard system maintained robust localization and perception at speeds up to 256.3 km/h and peak lateral acceleration of 26.8 m/s$^2$. In full-scale simulated racing, the well trained world-model-centric agent achieves an 88.3% interaction success rate across various challenging simulated racing scenarios. Closed-loop refinement of the world model and policy further improved utilization of cognitive-physical limits, recovery from failure modes, and generalization across varying conditions and unseen circuits. These results suggest a boundary-aware methodology in which world models help embodied agents represent, predict, and continually refine their capability boundaries for safer real-world deployment.
Read the full abstract
Discrete Mathematics
A Fine-Grained Complexity of Co-Secure Domination for Some Subclasses of Chordal Graphs
For a connected graph $G = (V, E)$, a set $D \subseteq V$ is a co-secure dominating set if $D$ is a dominating set of $G$ and for each vertex $u \in D$ there exists a vertex $v \in V \setminus D$ such that $uv \in E$ and $(D \setminus \{u\}) \cup \{v\}$ is a dominating set of $G$. In this article, we present bounds on the co-secure domination number (size of minimum co-secure dominating set) in some restricted 2-trees, a popular subclass of chordal graphs. We also show an interesting dichotomy that the co-secure dominating set problem is NP-complete on $K_{1,r}$-free split graphs, $r\ge 4$, whereas it is linear-time solvable on $K_{1,3}$-free split graphs.
Read the full abstract
Computation and Language
Simplex Relaxation for Discrete Diffusion
Discrete diffusion models for categorical generation are defined by a corruption kernel, which determines the intermediate state space and the associated reverse prediction problem. Within this family, uniform diffusion has been extensively developed, with recent work connecting its categorical corruption process to continuous representations and dynamics. Motivated by this view, we ask whether uniform discrete diffusion can be augmented with an explicit continuous state while leaving its categorical corruption process unchanged. We introduce Simplax, an exact Dirichlet–categorical augmentation that couples each corrupted categorical state with an auxiliary simplex-valued variable while preserving the uniform diffusion process as its categorical marginal. This augmentation yields a tractable Rao–Blackwellized reverse-bridge objective and a stochastic reverse sampler, while retaining the corrupted categorical state as the denoiser input. Empirically, Simplax improves the generative perplexity–entropy tradeoff on unconditional OpenWebText generation. On Sudoku, a model trained exclusively on $30$-clue puzzles achieves the highest accuracy among the compared methods across all evaluated clue densities, including the minimum uniquely solvable $17$-clue regime, and also achieves the highest validity in unconditional generation.
Read the full abstract
Cryptography and Security
Trigger the Straggler: Load Hijack on Mixture-of-Experts LLMs
Expert parallelism (EP) is a common strategy for serving large Mixture-of-Experts (MoE) models across multiple GPUs by distributing experts among devices. Router decisions then determine both which experts process each token and which GPUs execute the resulting work. This procedure exposes a supply-chain attack surface in the serving schedule. We introduce Load Hijack, in which a malicious model provider modifies only a checkpoint's router weights, distributes the poisoned checkpoint, and retains a private trigger. When the trigger appears, the poisoned router concentrates token-to-expert assignments on experts co-located on one GPU. The resulting load makes that GPU a straggler and forces peer devices to wait, while routing on ordinary inputs remains near the clean reference. We find this conditional behavior difficult to achieve because an objective that rewards target-expert use on triggered inputs can also bias ordinary-input routing toward the same experts. To resolve this conflict, Load Hijack employs a three-stage optimization procedure that produces strong trigger-dependent concentration while keeping ordinary-input routing close to the clean reference. Across three MoE families and four corpora, Load Hijack directs 92.3% to 95.6% of triggered token assignments to the target experts. In live EP serving, triggered traffic produces 1.43x the time-to-first-token and 0.86x the throughput measured under ordinary traffic. These results show that poisoned routers can act as trigger-controlled device schedulers and motivate checkpoint audits of routing and runtime load.
Read the full abstract
Software Engineering
CausalRepair: Bridging the Causality Gap in Large Language Model-Based Automated Program Repair via Dual-Slicing
Automated Program Repair (APR) has recently benefited from Large Language Models (LLMs), yet their effectiveness heavily depends on repair context. Existing LLM-based APR methods suffer from a causality gap: test contexts can be noisy or incomplete, while source contexts derived from static analysis often contain irrelevant and unexecuted code, misleading LLMs from identifying the true root cause. To address this issue, we propose CausalRepair, a conversation-driven APR framework based on minimal causal context, i.e., the essential dependencies required to explain a failure. CausalRepair employs a dual-slicing strategy: context-aware static slicing purifies test semantics, while execution-trace-based dynamic slicing captures precise runtime dependencies in source code. Together, they construct compact, causally relevant contexts to guide iterative repair. We evaluate CausalRepair on Defects4J V1.2, V2.0, and Defects4J-Trans using DeepSeek-V3. CausalRepair correctly fixes 313 bugs on Defects4J, outperforming state-of-the-art approaches such as ReinFix and TSAPR, while reducing the average repair cost to $0.029 per bug.
Read the full abstract
Networking
Extended Version: Measuring the End-to-End Resilience of Application Deployments in Real-World Communication Networks with DRACO
Billions of users take Internet connectivity for granted and use it daily to access applications deployed anywhere around the globe. What users perceive as a seamless connection between their end device and the application server involves a complex system of systems operated by various actors. Failures in any of these systems disrupt user-to-application connectivity if the communication network is not resilient. Since communication networks are a critical infrastructure as part of ICT, real-world operators rarely disclose information about them. This makes it difficult to measure the end-to-end resilience of deployments, because it requires guessing topologies and the availability of the involved systems. This paper presents DRACO, a framework for modeling communication networks and measuring end-to-end resilience of application deployments. It enables configuring application deployments across arbitrary communication networks and evaluating their availability, robustness, and resilience. We demonstrate the functional scope of DRACO by modeling nationwide application deployments in Germany and France, highlighting its compatibility with a wide variety of public data sources and synthetic data.
Read the full abstract
Numerical Analysis
A numerical approach for solving the time-Fractional Mobile-Immobile Transport Equation
Fractional diffusion models provide a powerful framework for describing anomalous transport phenomena in heterogeneous porous media. The Mobile-Immobile model is a fundamental approach for characterizing such anomalous diffusion, specifically addressing the delayed solute transport caused by mass transfer between mobile and immobile regions. In this study, we develop and analyze two fully discrete numerical schemes for the one-dimensional time-fractional Mobile-Immobile model governed by the Caputo derivative of order $\alpha\in (0,1)$. The spatial discretization is carried out using a non-symmetric interior penalty discontinuous Galerkin method, while the temporal derivative is approximated by the Crank-Nicolson L1 and L2-$1_\sigma$ formulas, resulting in two distinct schemes with high-order accuracy. Rigorous stability and error analyses are established, showing that the methods achieve optimal convergence rates depending on the regularity of the exact solution. Numerical experiments verify the theoretical predictions and demonstrate the efficiency and robustness of the proposed schemes. The presented framework provides a reliable and accurate approach for simulating complex fractional transport processes in porous media and related fields.
Read the full abstract
Computation and Language
ASR-Roundtrip Evaluation Can Mask Context- and Convention-Dependent Reading Errors in Chinese News TTS
ASR-roundtrip evaluation is widely used as a scalable proxy for text-to-speech (TTS) intelligibility, but it can produce false negatives for reading errors perceived by listeners. We study Chinese news TTS spans whose correct reading depends on context or domain conventions, such as sports scores, aircraft models, technical units, and membership names. In these cases, Raw TTS can choose a plausible but wrong reading while ASR transcribes the audio as the intended or surface-correct text. A targeted audit over 110 high-risk MiMo TTS cases, reported with a complete denominator, confirms 46 masked false negatives, 9 exposed TTS errors, and 55 cases with no Raw TTS error. A span-isolation diagnostic re-exposes 18/46 previously masked errors. A Raw-only CosyVoice audit on the same targeted pool confirms 51 masked cases. Across the 97 TTS-specific audio files labeled confirmed masked across the two audits, Qwen3-ASR surface-recovers 40 cases, whereas Paraformer does so in only 2. The results suggest that ASR-roundtrip is useful for screening but insufficient as standalone ground truth for Chinese news reading-risk evaluation.