Machine Learning
Papers
Fresh from arXiv, typeset properly. Highlighted words trade on the market.
Tuesday 18 August
Multi-step causal reasoning requires chaining inferences where each step constrains the next. An early error propagates silently, and a correct answer reached via flawed logic evades outcome-level detection. In specialized domains, teacher LLMs err on intermediate steps, safety constraints restrict cloud distillation, and shifting conditions demand adaptation, leaving self-evolution as the practical route. Naive self-evolution can collapse: outcome-only rewards let the model exploit distributional shortcuts, and weak self-evaluation reinforces spurious paths into stable failure patterns. We exploit a key asymmetry: generating a correct chain is hard, but verifying a single step is easy. Many high-stakes domains admit a deterministic, queryable oracle, a physics simulator or rule engine over codified constraints. It checks asserted steps without teacher-level ability and abstains beyond its rules; it can check what the model asserts, never replace it. This enables CoEvo, an oracle-grounded self-evolution framework where a single model alternates between Proposer and Solver. As Solver, the model generates competing chains; intra-group debate exposes disagreement steps, a proxy for the capability boundary, and the oracle adjudicates them into process-level supervision. As Proposer, the same model constructs progressively harder scenarios inside oracle constraints, steering the curriculum toward deep multi-hop chains. Both roles are updated jointly, so training pressure co-evolves with the model. On industrial, clinical, and legal multi-step causal reasoning benchmarks, CoEvo enables an 8B LLM to sustain self-evolution, surpassing distillation baselines and the strongest proprietary reference on path correctness (82.1% vs. 71.4%). The trained model generalizes to unseen categories and systems, preserving root-cause accuracy.
Computation and Language
Retrieved-Span Training for Efficient Query-Focused Meeting Summarization on QMSum
QMSum provides no scorer, making query-focused meeting summarization results difficult to compare. We rescore or generate 15 systems under one implementation. Through a common inference port, a released 406M Fusion-in-Decoder specialist loses 6.30 ROUGE-1 when moved from capped long input to 2,000-word retrieved spans. Fine-tuning it on this span regime recovers the loss. On test it scores 36.33 ROUGE-1 versus 35.41 for our 1.2B system; the meeting-cluster 95% interval for the difference is [-0.27, +2.22], so QMSum does not statistically separate them. The smaller system uses about one-third as many total parameters and less than half the peak inference memory. Within the fixed 1.2B base, span-regime fine-tuning adds 5.29 [+4.02, +6.56], while replacing the first 4,500 transcript words with 2,000 retrieved words adds 1.55 on test and 0.29 on validation. Separately, under one concise prompt and reference-overlap scorer, a released 406M specialist exceeds five proprietary hosted models by at least 6.2 ROUGE-1, but output length and absent human or factuality evaluation limit this ordering. Conclusions are limited to QMSum and automatic metrics.
Read the full abstract
Computation and Language
Privacy Personalization Trade offs in LLMs: The Impact of Stylometric Signal Reduction on User-Specific Text Generation
Large language models (LLMs) have demonstrated the ability to generate user-specific text with high stylistic fidelity. However, the personal data that enables such personalization frequently embeds demographic, cultural, and stylistic markers that raises concerns about stylometric re- identification. This paper investigates whether reducing identifiable stylistic signals affects personalization in text generation by LLMs. We introduce a controlled framework to isolate stylometric signals in LLM personalization using the LaMP-7 Twitter benchmark. Experiments on 250 sampled users compare two settings: paraphrasing conditioned on the original profile and paraphrasing conditioned on an anonymized converted profile in which demographic identifiers, cultural references, personal details, and informal linguistic cues have been systematically neutralized. Outputs are assessed by two independent LLM judges and a complementary human evaluation. Our pairwise evaluation shows that outputs conditioned on original profiles are nearly indistinguishable from human-authored ground truth, indicating that modern LLMs can closely reproduce an author's writing style with sufficient fidelity. In contrast, preference for model outputs with anonymized profiles drops to 13.0% on average, while semantic context preservation remains high at 94.8%. A study with human evaluators confirms the same pattern. These findings reveal a clear privacy-personalization trade-off and highlight the need for privacy-aware personalization methods that retain meaning while suppressing identifying stylistic signals.
Read the full abstract
Computation and Language
Beyond the Text: Verifying That Agent-Written Papers Are Backed by Their Artifacts
Large language model agents are increasingly capable of conducting research autonomously, producing research documents alongside the code and experiments that ostensibly support them. Yet whether the reported findings are consistently supported by corresponding implementations and execution evidence remains largely unexplored: existing review practices primarily assess textual quality and cannot reliably identify inconsistencies such as hard-coded metrics, unimplemented methods, or unsupported experimental results. We present ReAgent, an automated auditing framework for assessing the consistency between agent-generated research documents and their associated repositories. ReAgent constructs structured representations of scientific claims from research documents and uses them to guide repository analysis and evidence collection. Static auditing examines whether claimed methodologies, implementations, and experimental configurations are consistently reflected in the repository, while dynamic auditing executes relevant experiments and collects execution evidence to assess empirical findings. By combining static analysis with dynamic evidence, ReAgent identifies inconsistencies that may remain hidden under either perspective alone, such as experiments that reproduce reported numbers while deviating from the claimed methodology. The collected evidence and audit decisions are organized into a structured repository-level audit report, enabling transparent evidence traceability. We evaluate ReAgent on a manually curated benchmark of agent-generated research document–repository pairs and compare it against representative static and reproduction-based baselines. Experimental results demonstrate that ReAgent effectively identifies inconsistencies between reported research findings and their supporting repository evidence.
Read the full abstract
Machine Learning
A Learning Algorithm for Threshold Boolean Networks with Prescribed Fixed Points
We present a learning algorithm for inferring threshold Boolean networks (TBNs) with a prescribed set of fixed points. The proposed method employs a custom differentiable loss function that jointly enforces fixed point preservation, penalizes spurious attractors, encourages binary outputs, and promotes sparsity through L1 regularization. Applied to the FOS-GRN model of Arabidopsis thaliana, the approach achieved perfect reconstruction (i.e., all 10 desired fixed points and no spurious ones) in 5 out of 30 independent runs, recovering on average 8.53 $\pm$ 0.90 correct fixed points with no spurious attractors. In contrast, standard methods such as the Perceptron and Logistic Regression recovered up to 10 fixed points but introduced between 8 and 31 spurious ones. An additional analysis varying the sparsity coefficient ($\lambda$) confirmed that the method's performance and the structural properties of the inferred networks remain robust within a practical range (up to 0.01) of regularization strengths. Overall, the results demonstrate the effectiveness and stability of the proposed algorithm in capturing meaningful network dynamics under prescribed dynamical constraints.
Read the full abstract
Computational Engineering
Hemispherical Ray-Casting Analysis for Milling Configuration Classification and Machinability Assessment
Early-stage manufacturability assessment reduces design iterations and costly downstream changes. This paper presents a GPU-accelerated geometric analysis method that evaluates CNC machinability and setup requirements directly from CAD files. A multi-stage pipeline uses hemispherical ray-casting to quantify surface accessibility, followed by directional variability analysis to extract dominant machining axes. Threshold-based classification determines 3-axis versus multi-axis machining. For 3-axis parts, orthogonal ray-casting and reachability optimization compute the minimum setup count and explicitly flag inaccessible surfaces. Validation cases show computation times sufficiently short to support interactive CAD workflows, enabling quantitative manufacturability feedback during the design stage without requiring manufacturing expertise.
Read the full abstract
Graphics · Cross-listed from eess.SP
Lipschitz Extension Initialization for Moving Least Squares Reconstruction from Sparse Irregular Samples
The idea of using Lipschitz extensions [1,2], or Gradually Varied Functions (GVFs)[3], for mesh-free scattered data reconstruction was proposed by the author in 2012 [4]. However, its practical application to modern mesh-free reconstruction methods has not been fully explored. Motivated by recent advances in computational tools, including AI-assisted mathematical programming and software development, we revisit this idea and investigate the use of a Lipschitz extension as an initialization step for Moving Least Squares (MLS) reconstruction [5,6]. Our computational experiments indicate that this initialization significantly improves the stability and reconstruction accuracy of MLS under sparse and irregular sampling. This is a preliminary study intended to establish feasibility; a fuller evaluation with additional benchmarks and comparisons is left to future work.
Read the full abstract
Computation and Language
Can Large Language Models Identify Meaningful Touchpoints in Conversion Attribution?
Touchpoint selection in conversion attribution, namely identifying meaningful touchpoints contributing to conversions, is essential for e-commerce recommendation and online advertising. Current selection methods rely heavily on collaborative-filtering-based heuristics, which fail to align with user-perceived semantic intent. Through human annotation, we reveal a significant semantic gap: many implicitly-related, semantically relevant touchpoints remain undetected by existing rules. Therefore, we systematically evaluate the capability of Large Language Models (LLMs) in identifying these hidden associations. Our evaluation shows that while LLMs effectively uncover a substantial portion of implicitly-related touchpoints, significant room for improvement remains in their selection performance. Furthermore, we analyze the impact of different prompting strategies and foundation model choices on identification performance, providing valuable insights into their reasoning patterns and effectiveness. These insights offer a new roadmap for transitioning conversion attribution from mechanical rule-matching to human-aligned semantic reasoning. Moreover, we leverage the LLM-attributed conversion labels for enhancing industrial CVR model training and achieve significant offline performance gains, showing the potential of LLMs in conversion attribution.
Read the full abstract
Artificial Intelligence
Retrieving Relations, Detecting Fallacies: A RAG Approach to Political Debate Analysis
Fallacies are arguments that employ invalid reasoning, making their automatic detection critical in sensitive contexts such as high-stakes political debates, where public opinion is shaped. Spotting a fallacious argument requires contextual knowledge beyond its pure surface text. This entails world knowledge pertaining to the subject matter under discussion, as well as knowledge of the relationships that exist between arguments within the argumentative discourse. Prior work on fallacy analysis has shown that argumentative discourse structure can beneficially improve classification performance. However, such structure is typically encoded only as static classifier features, limiting its flexibility. Building on this intuition while addressing this limitation, we introduce a guided retrieval-augmented methodology for fallacy detection and classification that leverages argumentative relations of support and attack to dynamically steer the extraction of relevant documents. We evaluate our approach on the ElecDeb60to20 benchmark across 42 retrieval configurations and 14 models, performing retrieval over a 15GB knowledge base of collected political-related documents. Our approach improves macro-F1 up to 0.864 for fallacy detection and up to 0.725 for classification over non-retrieval baselines. These results show that incorporating external knowledge significantly enhances fallacy detection and classification when retrieval is argumentatively guided.
Read the full abstract
Systems and Control · Cross-listed from physics.chem-ph
Polyaniline as a conductive polymer and its role in improving the efficiency and conductivity of perovskite solar cells
This article investigates the role of polyaniline as a conductive polymer in the active layer of perovskite solar cells. Samples were created by incorporating polyaniline into the transport layers to assess its impact on enhancing efficiency and conductivity. The application of this polymer across various layers of the cell structure led to improved stability and performance. Given its high doping capability, polyaniline was examined in detail, particularly focusing on two types of oxidation doping and its integration into the hole transport layer. Graphene oxide and reduced graphene oxide were chosen as comparative models, and their performance was evaluated against the standard polyaniline configuration. Laboratory results revealed that power conversion efficiency increased by 17.5% with graphene oxide and by 36.8% with reduced graphene oxide. Furthermore, short-circuit current density improved by 9.8% and 23.1%, respectively. These findings are consistent with existing studies in the field and support the validity of the approach.
Read the full abstract
Robotics
Constructing Predictive Surgical Path for AI-based Capsulorhexis Skill Transfer
Automated training of surgeons is one of the most crucial factors that significantly minimize surgical training risks and expenses. With recent advances in artificial intelligence (AI) knowledge and available data from various surgeries, AI's involvement in surgical training is becoming very promising. It is recommended that at the early stages of AI development, it interferes in the surgery as a third agent alongside the trainer. As trust in AI increases, this process will lead to an AI agent acting as a trainer in the future. The first phase in which AI can intervene in the training process is to suggest an improved surgical path to the trainer. A platform must be constructed in the first step, to accomplish this task and to enhance the movement path of trainee surgeons. This paper introduces this platform along with an annotated capsulorhexis surgery dataset called the ARAS-Farabi dataset. In this research, a deep convolutional neural network is pre-trained with JIGSAWS and ARAS-Farabi surgical datasets that can extract surgical skill characteristics from surgery tool tip motion data. The proposed platform develops a reference model from the feature space of an expert surgeon's movement trajectory and proposes an improved path to enhance the skill of a novice surgeon. An optimization with two loss functions is utilized to create a path that raises the skill level of the novice surgeon's path while simultaneously predicting and preserving his/her intent. The results of this study reveal that, with the assistance of an AI agent, the trainee surgeon's movement path can be enhanced by at least 20 percent while maintaining his intentional objective. In addition to the recommended deep network, various tangible indicators have also been developed in this research to verify the level of trainee improvement.
Read the full abstract
Robotics
Geo-VLA: Geometry-Aware Vision-Language-Action Planning via Internalization of Map Semantics
Vision-language-action (VLA) models have advanced end-to-end autonomous driving by leveraging foundation models for semantic reasoning and long-tail generalization. However, their planning performance remains limited in complex driving environments because image-only representations inadequately capture planning-relevant road geometry and topology. In this paper, we propose Geo-VLA, a plug-and-play framework that enhances VLA models by learning geometry-aware visual representations. During training, Geo-VLA internalizes geometric map semantics to strengthen road-structure representations, while requiring no HD maps or additional lane information during inference. To support this approach, we introduce Geo-QA, a geometry-focused question-answering dataset that injects road geometry into vision-language representations through contrastive learning and instruction tuning. Experiments on NAVSIM v1 demonstrate that Geo-VLA consistently improves VLA planners with distinct action-generation architectures, achieving 92.1 PDMS and establishing a new state-of-the-art among single-camera VLA planners.
Read the full abstract
Computer Vision
WorldMind: Decoupled Game World Model for State-Aware NPC Behavior
Game world models have recently demonstrated promising capabilities in generating visually coherent and action-controllable gameplay videos. However, non-player character (NPC) behavior in existing models is either implicitly entangled with video generation or explicitly prescribed through external control signals. Consequently, a game world model has to jointly understand the state, plan the NPC's response and render its visual outcome, limiting its ability to produce responsive and state-aware NPC behavior. The challenge lies in the lack of an explicit interface for state-grounded decision-making. To this end, we introduce WorldMind, to our knowledge the first decoupled framework for state-aware NPC behavior in game world models. WorldMind separates interactive world modeling into four layers: an Understanding Layer that constructs a compact state from generated frames; a Decision Layer that reasons over the compact state to plan the NPC's next action; a Control Layer that translates the actions into temporally aligned conditions; and a Generation Layer that synthesizes their visual outcomes. By reconnecting layers in a closed interaction loop, WorldMind grounds NPC behavior in the evolving game state. We further introduce BOSS-140K, a dataset of gameplay videos paired with rich internal game states, together with an agent that automates the collection at scale. Experiments on BOSS-140K demonstrate reliable compact state reconstruction and mechanics-grounded planning, with WorldMind preferred over the baselines in approximately 70% of pairwise comparisons for its more tactically appropriate and coherent NPC behavior. Project page: https://teawhite.cn/worldmind_projectpage/
Read the full abstract
Computer Vision
DesignAgent3D: Interactive 3D Scene Editing via Designer-like Multimodal Reasoning
Text guided 3D scene editing provides an intuitive interface for modifying reconstructed environments, but remains difficult because natural language design requests are often semantically underspecified and must be grounded in cluttered 3D scenes. Existing methods typically formulate the task as one-shot conditional generation from a single prompt, failing to resolve ambiguous user intents or achieve precise spatial grounding. Consequently, they suffer from severe object localization drift, tracking failure under occlusions, and the notorious multi-view "sticker effect." To overcome these limitations, we present DesignAgent3D, an interactive multimodal agentic framework that reformulates 3D scene editing as a designer-like Plan-Perceive-Act paradigm. The agent first plans by interacting with the user to clarify underspecified design goals, then perceives by grounding the intended edit to specific objects or regions in the 3D scene, and finally acts by applying controlled visual modifications while preserving scene consistency. The edits are further integrated into the underlying 3D representation, supporting persistent and multi-view consistent novel-view rendering. Extensive experiments across both NeRF and 3D Gaussian Splatting backbones demonstrate that DesignAgent3D significantly outperforms state-of-the-art baselines, delivering superior semantic intent alignment, impeccable spatial localization accuracy, and high-fidelity multi-view consistency.
Read the full abstract
Artificial Intelligence · Cross-listed from q-bio.QM
Rigorous Evaluation of Large Language Models for Malaria Drug Discovery: Trade-offs in Performance, Scale, and Resource Utility
We introduce Malaria-Instruct, a curated instruction-following dataset derived from the ChEMBL Legacy Malaria corpus for Malaria virtual screening, and conduct a systematic evaluation of five open-source LLMs; Gemma-2 2B/9B, TxGemma-2B/9B, and LlaSMol-Mistral-7B, on a rigorous out-of-distribution data split. Performance was benchmarked against classical ML models (Random Forest, XGBoost) and frontier proprietary models (Gemini 2.5, OpenAI o3) under few-shot conditions. Fine-tuned LLMs substantially outperformed all baselines: TxGemma-9B achieved the highest ROC-AUC ($0.731 \pm 0.005$) and LlaSMol-Mistral-7B the best enrichment factor (EF@1% $\approx$ 4.99). Domain-specific fine-tuning proved categorically indispensable with TxGemma-9B collapsing from ROC-AUC 0.731 to 0.499, under its best few-shot condition, and neither Gemini 2.5 (ROC-AUC $\approx$ 0.53) nor o3 (ROC-AUC $\approx$ 0.59) achieved reliable discrimination without fine-tuning. Biomedical pretraining conferred a measurable advantage at equivalent scale, while chemistry-aware pretraining yielded superior prospective enrichment. Fine-tuned open-source LLMs represent a compelling, resource-efficient paradigm for antimalarial VS, outperforming both classical pipelines and proprietary reasoning models under structurally challenging conditions.
Read the full abstract
Computer Vision
Ultra-High-Definition Restoration Transformers with Correlation Matching Transformation
We propose UHDformer++, a general Transformer-based framework to solve numerous Ultra-High-Definition (UHD) image restoration tasks. UHDformer++ operates across $4$ coordinated learning spaces: 1) a high-resolution space (HR) for multi-level feature extraction, 2) a low-resolution space (LR) for learning compact, representative features, 3) a super-resolution space (SR) for upsampling low-resolution features from SR, and 4) a low-high fusion and reconstruction space (LHFR) for final image restoration. Specifically, HR extracts multi-scale high-resolution features and fuses them with low-resolution cues to produce residual images, while LR distills complementary representations from HR to improve restoration quality. To supply LHFR with richer features, SR super-resolves LR outputs before fusion. We further introduce two modules to bridge the high- and low-resolution spaces. The Feature-Refined Correlation Matching Transformation (FR-CMT) module selects the top $C/r~(C~\text{denotes the number of channels;~}r\geq1~\text{controls the squeezing level})$, from the fusion between max- and mean-pooled high-resolution features to replace less informative channels in the low-resolution Transformer. The Adaptive Channel Modulator (ACM) adaptively recalibrates multi-scale high-resolution features, ensuring that only task-relevant information propagates to LR. Extensive experiments demonstrate that UHDformer++ reduces model parameters by at least 86% compared with recent state-of-the-art methods while achieving substantial performance gains across $5$ UHD restoration tasks, including low-light image enhancement, dehazing, deblurring, deraining, and desnowing. Code will be released at https://github.com/supersupercong/uhdformerplus.
Read the full abstract
Computers and Society
Mapping General-Purpose AI Governance in Twenty AI Middle-Power Jurisdictions
The most capable general-purpose AI (GPAI) models are mostly built in two jurisdictions, the United States and China, but the risks they carry land globally. Regionally advanced economies hosting no frontier developer, which we call AI middle-powers, are writing their own rules to govern GPAI. This paper investigates which GPAI-relevant provisions these AI middle-powers have enacted, mapping twenty jurisdictions including the European Union at the level of the individual provision, across four governance areas that trace the accountability chain for the model layer: systemic risk assessment, evaluation and verification, prohibitions with monitoring and detection, and serious incident reporting. Confirmed absence is recorded as data alongside positive provision. We find that jurisdictions converge on form, but diverge on force. Sixteen engage in at least three of the four governance areas, yet only about one in five provisions sit in binding law, and three-quarters of the instruments that do bind do so without defining GPAI. The institutional infrastructure shows the same shape: four in five of the mapped governance actors hold mandates that predate GPAI, and obligations attach wherever the inherited regime already reached, which is the application layer rather than the model. Where these states engage the model layer, they build capacity to observe it rather than impose duties on those who build it, and almost every evaluation body was constituted without the power to act on what it finds. Nominal coverage of the full accountability chain reaches eleven jurisdictions, but only five hold more than one provision in every area and, outside the EU, no jurisdiction imposes a binding evaluation duty on a model developer. The dataset gives researchers and policymakers a provision-level basis for identifying where regimes could align, and where coordination would have to start from scratch.
Read the full abstract
Mathematical Software
DSLHyPE-a DSL kernel language for the Exascale Hyperbolic PDE Engine ExaHyPE
We introduce a bilingual domain-specific language (DSL) for modelling compute kernels within a generic solver for hyperbolic partial differential equations (PDEs). Users express PDE terms, i.e.~the underlying physics, in a familiar native language such as C or C++, while the numerical scheme is specified in a Python-embedded DSL, DSLHyPE. DSLHyPE's compiler lowers the Python description to MLIR and introduces a translation pass that integrates it with native code likewise mapped to MLIR. Our approach keeps the numerical representation and the physics implementation separate for as long as possible, while delegating optimization to the compiler through existing MLIR optimization passes. This separation of concerns benefits researchers developing numerical schemes on top of existing PDE implementations or with applications involving nonlinear systems whose PDE terms must solve PDEs themselves. We demonstrate the feasibility of the approach using a gravitational-wave solver and a matter-evolution solver on x86 processors and H200 GPUs.
Read the full abstract
Software Engineering
What Does an Evaluation License? A Commit-Bound Census of Claim Replay in Inspect Evals
Benchmarks can run without determining what their results license. We freeze a large evaluation collection and attempt to replay its historical claims. Most units stop because the evidence required for replay is not bound. Where replay is possible, different claims remain stable at different resolutions. We make this otherwise implicit inference step explicit and executable.
Read the full abstract
Cryptography and Security
Incident-Data Robustness Analysis of the OWASP Top 10 for LLM Applications (2026): How a Community-Expert Ranking Holds Up Against a Large-Scale LLM Incident Corpus
The OWASP Top 10 for LLM Applications ranks the risks that a community of security practitioners judges most important. We ask a narrower question: checked against the record of real incidents, does that expert ranking agree with the data? We assembled a large-scale corpus of LLM-security incidents (7,714 snapshotted and 6,639 labeled against the 20-entry taxonomy) drawn from CVE, GHSA, OSV, and AIAAIC, and derived an incident-based ranking with a Bayesian measurement-error model that corrects each category's count for classifier precision and recall. The 2026 candidate list blends the two signals at fixed weights, 0.75 on the expert vote and 0.25 on the data, so the corpus corrects the consensus without overturning it. The agreement between the two rankings is weak: Cohen's $\kappa \approx 0.20$, with a 90% interval that crosses zero. The expert ranking is nonetheless robust. A pre-registered bake-off of four frontier classifiers returns no winner. None beats the incidence floor's balanced accuracy of 0.863. A ground-truth check leaves the floor's ordering (Spearman $\rho = 0.918$ against held-out truth) in place. This is an exploratory analysis by two working-group members, not the official OWASP release, and it does not supersede the official list or process.
Read the full abstract
Software Engineering
The Evaluation Context Protocol (ECP): A Portable Contract for AI Agent Evaluation
The evolution of artificial intelligence has necessitated a fundamental shift from evaluating isolated Large Language Models (LLMs) to assessing autonomous agentic architectures. This paper explores the critical methodologies for evaluating AI agents and the essential role of advanced observability infrastructure. We analyze the architectural components of agents and identify the severe limitations of current evaluation paradigms, including benchmark exploitation, the "confidently wrong" phenomenon, and the discrepancy between theoretical capability and operational reliability. To begin addressing the fragmentation in current evaluation infrastructure, this paper proposes the Evaluation Context Protocol (ECP), an early-stage, vendor-neutral framework intended to act as a portable evaluation contract layer for agentic systems. In its current form ECP defines a small JSON-RPC interface over which an agent exposes its user-visible output, the tool calls it made, and evaluator-safe audit context, and against which programmatic checks can be run uniformly across frameworks and continuous integration systems. We describe an open-source reference implementation that includes adapters for LangChain, LlamaIndex, CrewAI, and PydanticAI, and we situate the design against failure modes documented in the recent literature. ECP is presented as work in progress rather than a finished standard: the evaluation surface, method set, and grader families are all expected to change as the protocol is exercised against more systems, and the empirical validation required to justify adoption is outlined as future work.
Read the full abstract
Systems and Control
LiDAR-Derived Surface Priors for Multimodal Sensing-Assisted NLoS Beam Search in Indoor 60-GHz Networks
Highly directional 60-GHz Internet-of-Things (IoT) links can exploit naturally occurring indoor surfaces to sustain connectivity under blockage. Identifying viable non-line-of-sight (NLoS) paths, however, can require extensive RF beam training. This paper investigates whether LiDAR can reduce this search overhead by providing a surface-aware prior without assuming a direct mapping between optical return and mmWave reflection. The proposed framework uses LiDAR-derived geometry and return statistics to rank candidate propagation directions, while RF measurements remain responsible for final beam selection. The experimental validation is organized in three stages to separate descriptor robustness, cross-modal association, and beam-search performance. Controlled LiDAR measurements first quantify how geometric and radiometric surface descriptors vary with acquisition geometry. Matched LiDAR and 60-GHz measurements in an L-shaped corridor then determine whether these descriptors are associated with the measured surface-mediated RF response under a prescribed NLoS interaction. Finally, a separate room-scale campaign evaluates the resulting prior using exhaustive TX-RX beam maps without prescribing the underlying propagation mechanism. The measurements show that descriptor reliability depends on acquisition geometry and point-cloud representation, and that LiDAR and RF surface responses exhibit cross-modal association without supporting deterministic RF-power prediction. In the room experiment, local three-ring 3-D planarity retains a beam within 3 dB of exhaustive search at 74.5% of the measured locations while reducing RF beam-pair probing by 72% relative to exhaustive probing over the candidate search region. These results establish LiDAR-derived local surface structure as a communication-oriented prior for concentrating RF probing and reducing mmWave beam-search uncertainty.
Read the full abstract
Information Theory
Quickest Change Detection in Parametric Models With 1-Bit Measurements
We consider the quickest change detection from 1-bit quantized observations, where the post-change distribution is a parametric model with unknown parameters and the quantization thresholds are jointly chosen with the detection statistic. We propose an Adaptive-Quantization-Threshold CUSUM (AQuTeCUSUM) algorithm, which estimates the post-change parameter online and adaptively selects the quantization threshold to maximize the induced Kullback-Leibler divergence. Under suitable regularity conditions, we characterize the average run length and worst-case average detection delay of the AQuTe-CUSUM procedure, and show that it is asymptotically optimal in first-order as the average run length goes to infinity. Finally, we assess the performance of AQuTe-CUSUM for two distributions, namely the Gaussian and Poisson.
Read the full abstract
Distributed Computing
SLO-Scaler: Uncertainty-Aware SLO-Driven Autoscaling for Microservices
Autoscaling microservice-based applications to satisfy Service Level Objectives (SLOs) remains challenging due to bursty workloads, cascading latency across service dependencies, and cold-start overhead. Existing approaches such as the Kubernetes Horizontal Pod Autoscaler (HPA) rely on threshold-based CPU or memory metrics, which react too slowly to traffic spikes. Recent predictive methods improve responsiveness but generate point forecasts that ignore prediction uncertainty, leading to over-provisioning or oscillatory scaling. We propose SLO-Scaler, an uncertainty-aware autoscaling framework that predicts short-horizon request rates, tail latency, and SLO violation probability using a Bayesian LSTM model. SLO-Scaler integrates confidence-interval-based scaling decisions with a dependency graph analysis module that localizes bottleneck services, avoiding unnecessary whole-chain scaling. We evaluate SLO-Scaler on the DeathStarBench Social Network benchmark deployed on Kubernetes under periodic, bursty, and long-tail traffic patterns. Under bursty traffic, SLO-Scaler reduces the SLO violation rate by 29-56%, lowers the average replica count by 18-33%, and decreases scaling event frequency by 38-59% compared with the baselines, while achieving lower tail latency.
Read the full abstract
Artificial Intelligence
Programmable Distributions of Semantics-Preserving Variants for Evaluating Code Agents
Code agents are increasingly deployed to resolve real software issues, yet their reliability under superficial code variations remains poorly understood. We evaluate whether coding agents that repair repository-level issues remain reliable when the surrounding codebase is rewritten into a semantically equivalent form. We sample semantically equivalent variants of a repository from an explicitly defined distribution and compare the agent's resolve rate on them with that on the original. Different distributions reveal different aspects of an agent, such as its sensitivity to routine refactoring or to misleading code near the location of a fix. We develop VDL, a probabilistic programming language for expressing such distributions, whose programs randomly apply semantics-preserving transformations (SPTs) to a repository. For our evaluation, we use it to define a generic distribution that is not targeted at any particular failure mode. We evaluate two agentic scaffolds (mini-SWE agent and OpenCode), each backed by one of four frontier models (Claude Opus 4.5, Kimi K2.5, MiniMax M2.5, and Qwen 3.6-27B), on instances from SWE-bench Verified and SWE-bench Pro. Resolve rates drop by up to 6.7 percentage points, significantly in 6 of 16 configurations. No ranking of models by robustness holds across scaffolds, revealing a jagged robustness frontier.
Read the full abstract
Computer Vision
Depth Anything V4: Dynamic 4D Scene Reconstruction via Riemannian Flow Matching on 4D Gaussian Splatting
We present Depth Anything V4 (DAV4), a framework for dynamic 4D scene reconstruction from monocular video. Our key contribution is the application of Riemannian Flow Matching (RFM) to 4D Gaussian Splatting parameters, defining probability paths directly on non-Euclidean manifolds (scale, rotation, opacity), ensuring all intermediate states are valid. Through controlled experiments, we isolate RFM's contribution from test-time optimization (TTO) and pre-training. A deterministic MLP baseline with the same data, architecture, and TTO achieves F-score 0.762; RFM achieves 0.806 - the +0.044 gain is RFM's isolated contribution. We provide corrected computational cost analysis: pre-training is 360 GPU-hours, amortizing for large-scale deployment (over 10,000 scenes). Uncertainty is quantified via Negative Gaussian Log-Likelihood and Expected Calibration Error. DAV4 outperforms prior Depth Anything models and per-scene 4D-GS on dynamic reconstruction and novel-view synthesis, while using no human-annotated depth labels as training losses.
Read the full abstract
Computer Vision
TTSD-FAR: Test-Time Self-Distillation with Fisher-Anchored Restoration for Missing-Modality Emotion Recognition in LVLMs
Large video-language models (LVLMs) have shown remarkable performance on multimodal tasks like multimodal emotion recognition (ER) in the wild. ER is inherently multimodal, requiring a joint understanding of facial expressions, vocalizations, language, biosignals, and gestures. However, real-world deployment remains challenging: modalities may be missing or noisy at test time. Partial observations can be viewed as a distribution shift relative to the complete-modality distribution. SOTA TTA methods based on entropy minimization or perplexity reduction do not transfer to autoregressive LVLMs, while retrieval augmented generation (RAG) degrades when the observed modality is weak. Because no ground-truth supervision exists to verify individual updates, adaptation across this stream risks accumulating drift and degrading once the model departs from a reliable solution. An effective solution must therefore adapt to arbitrary missing-modality patterns and remain effective during continual adaptation. We address both jointly with Test-Time Self-Distillation (TTSD), a parameter-efficient framework in which a frozen teacher, trained on complete modalities, guides an adaptive low-rank student via self-distillation, updating only a negligible number of parameters. Stability is built into this same loop through Fisher-Anchored Restoration (FAR), which monitors Fisher information stability to detect convergence versus drift and restores the student toward the teacher's anchor when distributional shifts are identified. Our experiments on MELD, DFEW, and BAH under 0%-50% missing modalities show that this unified adaptation-restoration design consistently outperforms entropy-based adaptation, RAG, and perplexity-based generation over long adaptation horizons, where baselines without restoration progressively degrade while TTSD-FAR remains consistent.
Read the full abstract
Numerical Analysis
Direction-Adaptive Plane-Wave Discontinuous Galerkin Methods for the Helmholtz Equation
We consider plane-wave discontinuous Galerkin (PWDG) approximations of the Helmholtz equation with adaptive local propagation directions. The directions are chosen by minimizing a weighted residual on the mesh skeleton. We study two formulations: in Part A the PWDG system is solved for the coefficients at fixed directions, while in Part B the same residual is minimized jointly over coefficients and directions, with the coefficients eliminated by variable projection. Complex angles unify propagating and evanescent Trefftz waves. The discrete system is normalized in the Trefftz-DG norm, and a local Cauchy-trace Gramian is used to remove numerically dependent directions. On straight edges the residual integrals are exact. For all-Dirichlet problems, the residual equals the squared DG error and provides local adaptive indicators. We prove local quadratic growth of the reduced direction functional near an identifiable zero-residual solution of fixed rank. Numerically, an exact circular DtN test shows that the dominant phase direction of a Hankel wave can be recovered accurately even when a small plane-wave fan gives a larger field error. We also separate the effects of the coefficient basis, trace cutoff, and trace-spectrum arithmetic on the high-$p$ DtN floor. Finally, for finite sums of plane waves, residual-based ENRICH–MOVE continuation recovers all directions to roundoff for $M=1,\ldots,19$. At $M=20$ the automatic birth step enters a false basin, whereas a nearby birth again reaches roundoff. The results indicate that direction adaptation is most effective for solutions of low directional complexity.
Read the full abstract
Machine Learning
Selection, Recombination, or a Fresh Solve? A Candidate-Free Control for Single-Pass Test-Time Aggregation
An aggregation pass gives a language model two things: candidate solutions and another chance to answer. We ask whether the candidates improve that attempt, and whether aggregation improves on selecting an existing answer. Our candidate-free control asks the same model to solve the problem again from the problem alone, using the same decoding settings and maximum output-token allowance. On AIME-2025 and HMMT-2025, Qwen3-4B-Instruct-2507 reaches 15.5% accuracy when all four candidates are wrong, compared with 27.7% for a fresh solve. When at least two candidates are correct, the ordering reverses: 88.7% for aggregation versus 59.7% for a fresh solve. With exactly one correct candidate, the benefit remains unresolved under the specified test. Yet plurality voting reaches 96.0% in the multiple-correct regime without another model generation; its accuracy difference from aggregation is inconclusive (raw $p = 0.180$). The all-wrong accuracy loss and multiple-correct gain survive a statistical correction for adding the second benchmark, developed after data collection, and persist when fresh responses are scored by their final answers. These findings concern one 4B model, two mathematics benchmarks, and a single aggregation pass over candidates truncated at their first answer. All-wrong recovery can arise from solving the problem afresh. The fresh-solve control tests whether candidates improve accuracy. Comparing with selection asks whether the additional generation improves on an existing answer.
Read the full abstract
Numerical Analysis
Fully discrete parameter-robust error analysis of a grad-div stabilized Crank-Nicolson artificial compressibility method for the Navier-Stokes equations
The artificial compressibility method (ACM) relaxes the incompressibility constraint in the Navier-Stokes equations by introducing a perturbation term proportional to the time derivative of the pressure, scaled by a small positive parameter $\varepsilon$. Consequently, the ACM system inherently involves two small parameters: the fluid viscosity $\nu$ and the artificial compressibility parameter $\varepsilon$. While existing temporal analyses of ACM provide insights into its behavior, rigorous fully discrete error estimates that are robust with respect to both parameters remain a significant gap in the literature. In this paper, we propose a second-order Crank-Nicolson fully discrete ACM scheme and establish its parameter-robust optimal error estimates. To enhance stability and ensure robustness, we incorporate two distinct grad-div stabilization strategies: one facilitates the decoupling of velocity and pressure computations, while the other guarantees robustness with respect to both $\nu$ and $\varepsilon$. For spatial discretization, we employ the Scott-Vogelius finite element pair, which is crucial for the decoupling and error analysis. The resulting parameter-uniform bounds are crucial for ensuring the long-time accuracy of the scheme, circumventing the exponential dependence on the Reynolds number typically introduced by Grönwall's lemma. Numerical experiments are provided to validate the theoretical findings and demonstrate the efficiency of the proposed methods.