# Research Papers — 2026-06 · bundle page 58 of 260 > Full-text markdown of papers #2851–2900 of 12993 published in 2026-06, newest first. AI-readable mirrors of the full 12993-paper 2026-06 archive (50 papers per file). Canonical page: https://aipod.dev/papers/archive/2026-06/page/58 [← Previous bundle](https://aipod.dev/papers/archive/2026-06/page/57.md) [Next bundle →](https://aipod.dev/papers/archive/2026-06/page/59.md) ### Graph-Based Modeling of Financial Volatility Dynamics Authors: Chuanzhen Wang, Alice Zhang, Wei Chen, Michael Brown Published: 2026-06-23 Categories: q-fin.ST, cs.CL, cs.LG arXiv: https://arxiv.org/abs/2608.26127v1 Accurate forecasting of realized volatility ($RV$) is crucial for risk management and derivatives pricing. Although the implied volatility ($IV$) surface offers rich informational content, prevailing methods that treat it as a static image fail to capture its inherent dynamics. To overcome this limitation, we propose the Finance-Aware Graph Spatio-Temporal Network (FA-GSTN), a novel architecture that reframes $RV$ forecasting as modeling the evolution of a structured financial object. FA-GSTN builds a spatio-temporal graph sequence from the $IV$ surface, where nodes correspond to grid points and edges encode adaptive spatial (intra-day) and explicit temporal (inter-day) dependencies. The model incorporates domain knowledge through finance-aware node features (e.g., option Greeks) and tackles high-frequency noise via a multi-scale temporal smoothing gate coupled with an adaptive robust loss function. Comprehensive evaluations on a large-scale equity options dataset show that FA-GSTN sets a new state of the art, delivering superior predictive accuracy ($R^2$ up to 0.473). It also demonstrates remarkable data efficiency, substantially outperforming strong Vision Transformer baselines when trained on only one year of data ($R^2$: 0.372 vs. 0.315). Furthermore, the model exhibits enhanced robustness during periods of market stress, such as 2020--2021. Ablation studies confirm the vital roles of the spatio-temporal graph structure, finance-aware components, and integrated noise-handling modules. Our work underscores the substantial benefits of explicitly modeling temporal dynamics and infusing financial inductive biases for accurate and robust volatility forecasting. --- ### VA-DPO: Valence-Arousal Direct Preference Optimization for Controllable Emotion Generation in Language Models Authors: Hyunwoo Kim Published: 2026-06-23 Categories: cs.CL, cs.AI, cs.LG arXiv: https://arxiv.org/abs/2608.20374v2 How precisely can we tell a language model how to feel? Most work on emotional generation answers with a discrete label - happy, angry, sad - which cannot express a target like "mildly downcast but calm." We instead specify the desired affect as a continuous point (v*, a*) in the Valence-Arousal plane and train the model to hit it. Our method, VA-DPO, is a small modification to Direct Preference Optimization: a frozen VA regressor scores each sampled generation by its Euclidean distance to the target, we keep only candidate pairs whose distance gap clears a margin tau, and we optimize a LoRA adapter with the ordinary DPO loss against a frozen reference. The DPO objective itself is unchanged; what is new is how the preference data is built. On Llama-3.1-8B-Instruct this cuts mean VA distance to the target by 33% over system-prompting and 25% over few-shot prompting, lifting valence/arousal correlation to r_v=0.93 and r_a=0.75. The gains carry over to Qwen3-8B and Llama-3.2-3B, and they do not come at the usual price: MMLU is unchanged (Delta=+0.0) and HellaSwag and TruthfulQA are preserved. We release the code, configs, and the preference-construction pipeline. --- ### An ambiguity taxonomy for evaluating large language model performance on clinical registry abstraction: a multi-site prospective study Authors: James Matheson, Betsy Castillo, Andrew Y. Shin, David Scheinker Published: 2026-06-23 Categories: cs.CL arXiv: https://arxiv.org/abs/2608.20373v2 Objective: To evaluate large language model (LLM) performance on unprocessed electronic medical record (EMR) data for clinical registry abstraction. Methods: We evaluated LLM performance answering registry questions for the American College of Cardiology National Cardiovascular Data Registry (ACC NCDR). In a pilot study at an academic medical center, the model identified candidate data sources for each registry question and experienced abstractors used these results to define question-specific document sets. In a validation study at a second center with a second ACC NCDR registry, the LLM answered questions using the question-specific document sets. Before reviewing any output, two abstractors independently established the ground truth and assigned each question to one of six categories, ordered by the ambiguity and clinical reasoning required to resolve it: Medication/Event Flag, Binary Clinical Presence, Administrative, Quantitative Laboratory/Physiologic, Clinical Interpretation, and Event Timing. Results: The analytical sample comprised 9,430 abstractor answers reconciled to 4,715 consensus answers (501 pilot; 4,214 validation). In the pilot, candidate data sources per question averaged between 14.6 (SD 13.9) for demographics and 89.2 (SD 56.1) for history and risk factors. In validation, human inter-rater agreement was approximately 98\% while 87\% of LLM answers exactly matched consensus, 2\% partially, and 9\% did not. Mean question-level accuracy was 91.5\% (SD 13.4\%) across 157 questions with at least 20 answers, and declined as ambiguity increased, from 96\% for Medication/Event Flag to 62\% for Event Timing questions. Conclusions: LLMs answering clinical registry questions on unprocessed EMR data achieved far lower accuracy than human abstractors. LLM accuracy fell steadily as ambiguity and the level of required clinical reasoning increased. --- ### Edge-Based Agentic Retrieval-Augmented Generation for Autonomous FHWA Bridge Inspection Compliance Authors: Viraj Nishesh Darji, Hemaliben Rakeshkumar Darji Published: 2026-06-23 Categories: cs.IR, cs.AI, cs.MA arXiv: https://arxiv.org/abs/2608.20372v1 The Federal Highway Administration (FHWA) mandates that over 600,000 bridges in the United States be evaluated against the Recording and Coding Guide for the National Bridge Inventory (NBI). Manual compliance verification is labor-intensive, error-prone, and impractical in connectivity-limited field environments. This paper introduces BridgeGuard, a fully air-gapped agentic Retrieval-Augmented Generation (RAG) system for autonomous bridge inspection compliance. BridgeGuard integrates vector search over the FHWA Recording and Coding Guide with structured SQL queries against NBI tabular data, orchestrated by a stateful multi-step ReAct planning loop executing locally on commodity edge hardware. A section-aware chunking algorithm preserves hierarchical regulatory item boundaries, achieving 94.2% chunk integrity compared with 28.4% for naive fixed-size splitting. Evaluated on the full Delaware 2023 NBI inventory (874 bridges) and a Texas sample (200 bridges), the system achieves 99.77% and 100.0% classification accuracy, respectively, for Structurally Deficient bridge identification, with 100.0% citation accuracy, at 197.0 bridges per hour with out external network access. Ablation experiments confirm that both vector search and the multi-step agentic loop are necessary for correct compliance reasoning. --- ### Solving Is Not Drawing: A Benchmark for Diagrammatic Reasoning in Olympiad Geometry Authors: Hsien Xin Peng, Anthony Kim, Alvin Li, Calvin Supasanya, Shivank Garg, Kevin Zhu Published: 2026-06-23 Categories: cs.AI arXiv: https://arxiv.org/abs/2608.18111v1 Foundation models such as GPT and Claude now solve olympiad-level mathematics with remarkable proficiency, so much so that geometry problem solving has become a standard proxy for their mathematical reasoning. Yet solving a geometry problem and drawing the figure it depends on are not the same skill: progress often hinges on a faithful diagram with the right auxiliary constructions and incidences, and it is unclear that a model which reasons its way to the answer can also produce one. A growing collection of benchmarks, including MathVista, and MathVerse, measures whether models reach the correct answer, but to our knowledge, none isolate the distinct ability to construct the diagram itself, leaving this capability unmeasured. We introduce an open-source benchmark that targets this gap: 954 self-contained olympiad geometry problems, with a 297-problem hard subset, each paired with its solution and a human-authored, high-fidelity diagram in renderable Asymptote code, together with a suite of text-, code-, image-, VLM-, and constraint-based metrics for what we term diagrammatic reasoning. Evaluating current foundation models reveals a pronounced gap between solving and drawing: their diagrams are markedly less faithful, with an average compile success rate of only 36.14\%. Strong mathematical reasoning, we find, does not imply the ability to construct accurate geometric diagrams. Our benchmark and dataset can be accessed at https://huggingface.co/datasets/max98765/hard_geometry_problems_with_diagrams. --- ### When Chain-of-Thought Helps and When It Hurts: An Empirical Investigation of the Serial-Depth Bottleneck in LLM Reasoning Authors: Tughanbulut Kurtulush Published: 2026-06-23 Categories: cs.CL, cs.AI, cs.LG arXiv: https://arxiv.org/abs/2608.09942v1 It is widely assumed that chain-of-thought (CoT) prompting universally improves LLM reasoning. We investigate this through the conceptual framework of the H_dp bandwidth bound (Chen et al., 2024): although the formal bound binds only asymptotically (at astronomically large prompt lengths), it identifies a real architectural bottleneck -- serial computation exceeding a transformer's single-pass capacity must be externalised, which is what CoT does. Our central finding is a within-benchmark serial-depth gradient: single-pass (no-CoT) accuracy degrades monotonically with per-item serial depth, while CoT is approximately depth-invariant. We measure CoT effects across three instruction-tuned models (Qwen-2.5-7B/32B, Llama-3.1-8B) and five standard NLP benchmarks at practical context lengths. On high-depth P-complete tasks (GSM8K, MATH), CoT gives a +54 to +68 pp recovery gap across all models. On shallow TC^0 tasks (MMLU, ARC), CoT is structurally redundant (Delta in (0.0, +4.6) pp, no significant negative effect) -- though high no-CoT baselines (up to 95% on ARC) may reflect contamination, so this null is not a clean architectural test. The intermediate class L (HumanEval) shows a model-size-dependent transition: +23.2 pp (32B), +9.1 pp (8B), -28.7 pp (7B). The cross-benchmark depth-recovery correlation is Spearman rho = 0.661 (p = 0.007, n = 15); 9 of 15 benchmark-level McNemar tests are significant after Bonferroni correction. Pre-registered on OSF, our results indicate that CoT is not a universal reasoning enhancer but acts as a bandwidth bypass: it helps serial computation that strains single-pass capacity and is redundant for tasks that already fit. --- ### Innovating with Generative AI: A Human Bottleneck Framework Authors: Julian De Freitas, Ayelet Israeli, Gideon Nave, Artem Timoshenko, Olivier Toubia Published: 2026-06-23 Categories: cs.HC, cs.AI, econ.GN arXiv: https://arxiv.org/abs/2608.07504v1 We propose a human bottleneck perspective for understanding how generative AI transforms the innovation process. The central premise is that many constraints traditionally plaguing the innovation process are cognitive and social in origin, rooted in how people generate ideas, evaluate novelty, and communicate through social systems. Generative AI does not act uniformly on these constraints. At each stage, it can deepen some bottlenecks while alleviating others, and predicting these outcomes requires understanding the underlying mechanisms of the constraint itself. We identify bottlenecks in four stages of the innovation process: ideation, screening and testing, preference measurement and consumer insight, diffusion, and market learning. By grounding analysis in human behavior rather than rapidly changing AI capabilities, we offer a framework for assessing whether new developments alleviate or intensify the bottlenecks that matter most at each stage. We also distinguish bottlenecks likely to narrow as capabilities improve from those rooted in enduring human constraints. We further discuss AI-related issues that cut across the entire innovation pipeline, challenging the very existence and structure of the traditional innovation process. --- ### Governing Execution Risk in Agentic AI Systems: A Trajectory-Guided Framework for Red Teaming Authors: Zhihao Zhu, Yi Yang Published: 2026-06-23 Categories: cs.CY, cs.AI arXiv: https://arxiv.org/abs/2608.04018v1 AI agents are increasingly embedded in organizational workflows, where they interact with external information sources and invoke digital tools to perform operational tasks. As organizations adopt such systems, a critical challenge is identifying and mitigating risks arising from malicious or untrusted external information that can steer agents toward unintended actions. Existing red-teaming approaches largely rely on fixed attack templates or final attack outcomes, providing limited visibility into how attacks unfold through multi-step reasoning and tool use. We argue that agent execution risk should be understood as a trajectory-level phenomenon. Building on this perspective, we propose TrajRed, a trajectory-guided red-teaming framework that uses execution trajectories to uncover vulnerabilities in agentic AI systems. We further develop TrajGuard, a runtime governance layer that uses high-risk trajectories discovered during red teaming to monitor and intervene in ongoing workflows. Experiments on AgentDojo across four organizational task suites show that TrajRed identifies substantially stronger vulnerabilities than fixed-template and automatic red-team baselines. Building on these vulnerability findings, TrajGuard reduces attack success across all evaluated attack methods to near zero while preserving benign task utility. Together, the results demonstrate that execution trajectories provide a practical foundation for both red teaming and risk control in agentic AI systems. This work highlights the importance of governing agent execution in organizational AI deployments. --- ### AI-driven Multimodal Representation Learning for Latent Mediation Structure Discovery of Socioeconomic Disadvantage, Psychosocial Factors, and Cardiometabolic Multimorbidity: Insights from the All of Us Research Program Authors: Cong Cao, Shuangge Ma Published: 2026-06-23 Categories: cs.CY, cs.AI, cs.LG, stat.AP arXiv: https://arxiv.org/abs/2608.04016v1 Social disadvantage is associated with multimorbidity, but the pathways linking social conditions to disease burden remain poorly understood. We developed an AI-driven multimodal mediation framework that integrates socioeconomic, psychosocial, clinical, laboratory, behavioral, and genomic data from the All of Us Research Program. Modality-specific variational autoencoders were used to derive latent representations of each data domain, and mediation analyses were subsequently performed in latent space to evaluate indirect associations between socioeconomic disadvantage, psychosocial factors, and multimorbidity. The final analytic cohort included 20,804 participants with complete multimodal data. Across 800 exposure--mediator--outcome combinations, mediation signals were concentrated within a small number of latent dimensions. The strongest indirect association linked a socioeconomic disadvantage dimension, a psychosocial vulnerability dimension, and a cardiometabolic multimorbidity dimension (NIE = 0.002517). The psychosocial dimension was characterized by poorer mental health, greater loneliness, lower social well-being, and lower health literacy, whereas the outcome dimension was associated with hypertension, diabetes, hyperlipidemia, obesity, chronic kidney disease, and heart disease. Bootstrap analyses supported the stability of the leading pathway. These findings suggest that psychosocial vulnerability was strongly represented in the dominant latent pathway linking socioeconomic disadvantage and cardiometabolic multimorbidity. More broadly, the proposed framework illustrates how AI-based representation learning can be used to investigate complex relationships across high-dimensional multimodal health data. --- ### AI Security Priorities: A Field-Wide Agenda Authors: Gil Gekker, Rachel Steratore, Everett Smith, Asher Brass-Gershovich, Varun Gandhi, Nicole Nichols, Vijay Bolina, Buck Shlegeris, Lisa Einstein, Dan Lahav, Omer Nevo, Sella Nevo Published: 2026-06-23 Categories: cs.CY, cs.AI, cs.CR arXiv: https://arxiv.org/abs/2607.26069v1 As AI systems are rapidly integrated into critical economic, governmental, and national security functions, the gap between AI adoption and AI security readiness continues to widen. This paper presents a prioritized agenda for advancing AI security, informed by structured interviews with leaders across industry, government, and civil society, and refined through a multi-sector expert workshop. Participants identified and ranked the highest-importance and most cost-effective areas where progress could strengthen AI security - from protecting frontier AI systems and their underlying infrastructure to improving cybersecurity practices as AI reshapes the threat landscape. The resulting priorities are organized across four themes: establishing strategic foundations and policy frameworks; advancing public-private coordination and institutional infrastructure; advancing technical security engineering and assurance; and governing agentic AI under adversarial pressure. For each priority area, expert authors provide detailed analyses that define the problem, assess the current landscape, and identify actionable projects that stakeholders across sectors can pursue. The paper aims to serve as an initial practical foundation for coordinated investment and action across the AI security field. It is designed to serve both current practitioners and individuals and organizations looking to enter the field by identifying concrete, high-impact contributions suited to a range of strengths and capacities. --- ### The Human Utility Factor: A Computable Welfare Metric That Reframes AI Governance as a Constrained Optimisation Problem Authors: Sivasathivel Kandasamy Published: 2026-06-23 Categories: econ.GN, cs.AI, cs.CY arXiv: https://arxiv.org/abs/2607.26068v1 Existing AI governance frameworks, including the EU AI Act and NIST AI RMF, address safety, transparency, and accountability but do not operationalize quantitative constraints on macro-socioeconomic stability. As a result, AI systems may satisfy regulatory requirements while contributing to labor displacement, rising inequality, and reduced economic resilience. We introduce the Human Utility Factor (HUF), a differentiable welfare metric that models the interaction between Agency, Wellbeing, and Economic Stability as functions of three actionable policy levers: automation depth, redistribution intensity, and employment coverage. HUF yields a closed-form optimal automation level and a minimum redistribution threshold below which no level of automation is welfare-positive, transforming high-level governance objectives into computable constraints. We evaluate HUF using a three-agent multi-agent reinforcement learning framework across U.S., Canadian, and Nordic policy regimes. Both analytical and PPO-based agents identify welfare-optimal operating regions and reveal a critical failure mode: welfare metrics that do not explicitly constrain redistribution can converge to high-automation equilibria that satisfy the metric while undermining its intended societal objectives. Our results suggest that AI governance is fundamentally a constrained optimization problem rather than a compliance exercise. HUF provides a quantitative framework for evaluating automation policies, identifying socioeconomic stability boundaries, and supporting governance decisions under accelerating AI deployment. --- ### AutoCluster, AutoTopicModeling, AutoTrendAnalysis: A Complete AutoML Pipeline for Predicting Emerging Trends Authors: Ahmed Abolfadl, Marwa Mahmoud Abla, Mervat Abu-Elkheir, Maggie Mashaly Published: 2026-06-23 Categories: cs.DL, cs.AI, cs.LG arXiv: https://arxiv.org/abs/2607.22641v1 Predicting emerging trends is vital for businesses, researchers, and policymakers; yet traditional approaches often lack scalability and adaptability. This paper presents a trend prediction framework based on Automated Machine Learning (AutoML), designed to extract insights from textual datasets with temporal attributes. The system ingests subject-specific textual entries accompanied by a date field. The pipeline begins with preprocessing and embedding, followed by AutoClustering, which uses meta-learning to select the optimal clustering algorithm. AutoTopicModeling then applies successive halving to identify the best topic modeling method: Latent Dirichlet Allocation (LDA), Latent Semantic Analysis (LSA), BERTopic, or Non-negative Matrix Factorization (NMF) based on the coherence score for each cluster. For trend forecasting, AutoTrendAnalysis evaluates multiple models: Facebook Prophet, AutoRegressive Integrated Moving Average (ARIMA), Seasonal-Trend decomposition using Loess (STL), and Long Short-Term Memory (LSTM) selecting the most accurate based on Root Mean Square Error (RMSE), either through successive halving or exhaustive comparison. Topics are classified as strong signals, weak signals, or noise based on forecasting outcomes, enabling the identification of emerging trends. By automating clustering, topic modeling, and time series forecasting, this research enhances trend prediction accuracy while reducing manual effort. The proposed system offers a scalable and user-friendly solution suitable for real-time applications and stakeholders with limited machine learning expertise. Experimental results demonstrate that the proposed system's best trial achieves a final RMSE of 7.099, indicating high predictive accuracy. --- ### AI-Assisted Causal Inference and Mediation Analyses of Environmental and Psychosocial Determinants of Subjective Cognitive Difficulties in the All of Us Research Program Authors: Cong Cao, Shuangge Ma Published: 2026-06-23 Categories: cs.CY, cs.AI, cs.LG arXiv: https://arxiv.org/abs/2607.22640v1 Short-term environmental exposures have been linked to cognitive and behavioral outcomes, although many reported associations may reflect broader geographic and contextual differences. Using longitudinal data from the All of Us Research Program (2018--2024), we linked daily weather and air-pollution exposures to repeated attention-related and subjective cognitive outcomes. Associations were evaluated using pooled, fixed-effects, lagged, and event-study analyses. Additional machine-learning analyses were conducted to explore potential heterogeneity and latent psychosocial structure. Replication analyses were performed using the 2024 Behavioral Risk Factor Surveillance System (BRFSS). Several environmental exposure measures showed small associations with cognitive outcomes in pooled analyses, but most attenuated substantially after accounting for within-location temporal variation. Mediation, sensitivity, and machine-learning analyses yielded similar conclusions. In contrast, mental-health burden, loneliness, and social functioning were consistently associated with subjective cognitive difficulty and exhibited substantially larger effect sizes than environmental exposures. Similar patterns were observed in BRFSS. Exploratory AI-assisted analyses yielded findings broadly consistent with the primary longitudinal analyses. These findings suggest that short-term environmental perturbations may have limited associations with cognitive outcomes after accounting for within-location variation, whereas psychosocial factors appear to be more consistently associated with subjective cognitive burden. --- ### Inducing Comparability of Factorised Probability Distributions Authors: Jan Speller, Malte Luttermann, Marcel Gehrke, Tanya Braun Published: 2026-06-23 Categories: cs.AI arXiv: https://arxiv.org/abs/2607.20502v1 To allow for principled comparison between two probabilistic graphical models defined over non-identical variable sets, they have to be lifted to a common measurable space. To this end, we propose an extension scheme for any two given models and establish the formal foundation: Unmatched components are completed using conditionally uniform (Laplace) extensions such that the resulting joint distributions differ from the original ones only by multiplicative constants and coincide under projection. This preserves the probabilistic semantics while enabling the application of well-defined distributional discrepancy measures. We establish the invariance of the induced joint under projection and use the extensions to provide a minimal structural extension of two factor graphs to the smalles common measurable space as well as to a common graphical structure by a deterministic algorithm. In addition, we discuss structural and measure-theoretic properties and identify promising criteria for comparison methodologies. --- ### Operator-Aware Mixed-Precision Tolerance Calibration for Tensor Kernels Authors: Dipankar Sarkar Published: 2026-06-23 Categories: cs.LG, cs.MS arXiv: https://arxiv.org/abs/2607.16228v2 Most tensor-kernel correctness tests go through a fixed-shape all close-style check with hand-picked absolute and relative tolerances. The thresholds are copied across the corpus and rarely revisited. We mine the element-wise error distribution of every test case from accumulated cloud GPU runs across the 26-entry gpuemu corpus and 2 dtypes (8,076 result rows). We then ask one empirical question: what absolute tolerance would the kernel itself, observed under its correct implementation, justify? The answer is much tighter than the current hand-picked atol. The largest tightening is attention_triton fp16 at $2{,}184\times$. Restricted to the seven LLM-style buggy variants for which the corpus ships a paired correct counterpart, calibrated per-(op, dtype) tolerances raise bug-detection recall from 73.2% (1,805 of 2,467) to 82.4% (2,034 of 2,467), an absolute gain of 9.3 percentage points (+229 new detections). The control false-positive count rises from 0 to 20 out of 1,882 correct-control cases (+1.1 percentage points). --- ### LLM Unlearning for Cyber Defense: A Survey on Methods, Challenges, and Emerging Threats Authors: Ruppikha Sree Shankar, Abhishek Bhardwaj, Arnav Doshi, Anusri Nagarajan, Troy Paulus Asia, Saptarshi Sengupta Published: 2026-06-23 Categories: cs.LG, cs.CR arXiv: https://arxiv.org/abs/2607.16227v1 LLMs are increasingly deployed in security-critical systems across healthcare, finance, education, and decision support, yet their inability to forget creates serious cybersecurity, privacy, and safety risks. Sensitive personal information, copyrighted material, hazardous domain knowledge, and memorized training data remain encoded across billions of parameters long after deployment, leaving models vulnerable to extraction, jailbreak attacks, membership inference, and regulatory non-compliance. Real-world incidents, from chatbots regenerating private information to fabricated legal citations producing direct legal and financial cost, place the problem at the center of the emerging-threats landscape rather than the realm of speculation. Because retraining billion-parameter models on revised corpora is computationally infeasible, and because knowledge within an LLM is distributed and entangled across parameters rather than localized to identifiable units, LLM unlearning has emerged as the principal cyber defense response, aiming to remove or suppress targeted knowledge from a trained model without retraining and without eroding what the model should still know. A central question, however, remains unresolved. Do current methods genuinely remove knowledge, or do they only stop the model from expressing it under ordinary prompting conditions? This survey examines LLM unlearning through the lens of security, robustness, and verifiable forgetting, with primary focus on gradient-based methods, which have come to dominate the field due to their compatibility with existing training pipelines and their scalability to billion-parameter models. --- ### ERP Data Provisioning Financial Control Testing Authors: Anitha Samudrala Published: 2026-06-23 Categories: cs.LG arXiv: https://arxiv.org/abs/2607.09712v1 Financial control testing increasingly depends on representative enterprise resource planning (ERP) data in quality environments, yet direct production copies expose personal, supplier, banking, and commercially sensitive records. This work presents Secure ERP Quality Provisioning for Financial Control Testing (SEQ-FCT), a governed data-provisioning framework that combines deterministic masking, synthetic scenario expansion, referential tokenization, policy-based release approval, and automated validation for reconciliation, fraud-rule testing, and audit analytics. A single synthetic dataset is used for evaluation. It contains 186,000 finance-process records from six subsidiaries over 2022-2025, including accounts payable invoices, payments, general-ledger journals, accounts receivable receipts, and bank-statement lines. The dataset includes entity relationships, monetary values, approval paths, tax attributes, banking markers, exception labels, fraud-rule triggers, and control-failure outcomes. Because the dataset is synthetic, reported results demonstrate controlled internal consistency rather than production validation. Against a production-clone upper bound, static masking, rules-only synthesis, conditional tabular generative synthesis, and a hybrid baseline, SEQ-FCT achieved 0.932 reconciliation F1, 0.887 fraud-trigger recall, 0.914 control-failure F1, and an estimated leakage-risk score of 0.018. The analysis indicates that financial process behavior can be preserved more reliably when masking, synthetic data, and governance checks are evaluated as a single release pipeline instead of independent utilities. --- ### EvoClawBench: Can Agents Learn Reusable Skills from Their Own Runs? Authors: Zhiyuan Peng, Xin Yin, Chenhao Ying, Zhe Cui, Zixiang Ding, Zhenhua Liu, Jiang Wu, Yuan Luo Published: 2026-06-23 Categories: cs.LG, cs.SE arXiv: https://arxiv.org/abs/2607.09711v1 Existing agent benchmarks primarily test task completion, tool use, or skill utility, but do not isolate whether a runtime can convert evidence from its own runs into reusable skills that improve fresh executions after authoring overhead. We introduce EvoClawBench, a benchmark for this closed-loop skill-learning question on repeated, fixture-backed tasks. EvoClawBench compares direct execution without skills, PreSkill authoring before execution, and PostSkill summarization from first-run evidence followed by a fresh second execution. The suite contains 100 tasks and 502 sub-problems across coding, data, office, security, operations, and domain-document workflows, with support for multiple agent runtimes. Experiments with OpenClaw and nanobot under local execution show that direct baseline performance is strongly runtime-dependent: OpenClaw remains below 20% across models, while nanobot ranges from 56.45% to 96.13%. Self-authored skills have mixed effects. nanobot GPT-5.4 stays above 96% in all modes and MiniMax-M2.7 improves from 90.97% to 94.50% under PostSkill, but nanobot DeepSeek-V4-Pro drops from 77.77% to 4.80% with PreSkill and 0.99% with PostSkill. OpenClaw shows similarly non-monotonic behavior, with some skill runs near baseline and others collapsing. These results indicate that learning reusable skills from an agent's own runs is selective and cost-sensitive, rather than an automatic benefit of adding skill authoring to an agent loop. --- ### Hyperbolic Manifold Constrained Tabular Neural Network Authors: Tian Li, Huizhi Liang, Lucy Robinson, Varun Ojha Published: 2026-06-23 Categories: cs.LG, stat.ML arXiv: https://arxiv.org/abs/2607.09710v2 Tabular prediction is central to a wide range of real-world applications. Tabular data typically contain heterogeneous features as well as rich and complex relational information that can imply a latent structural manifold. Hyperbolic geometry can help capture complex structural relations in data. However, most existing tabular prediction models are constructed and optimized in Euclidean space. How to incorporate hyperbolic geometry into supervised tabular learning remains underexplored. We propose \textbf{HTNN}, a supervised hyperbolic manifold constrained tabular neural network for tabular prediction. HTNN consists of a hyperbolic feature-value representation layer for heterogeneous categorical and numerical features, followed by a conventional MLP predictor. HTNN employs a \emph{geometry-aware training} and \emph{geometry-free inference} optimization framework. The \emph{geometry-aware training} allows hyperbolic geometry to shape the latent representation learning of heterogeneous feature values. After training, the latent hyperbolic representations can be converted into ordinary Euclidean space for efficient \emph{geometry-free inference}. We conducted extensive experiments on the TALENT benchmark. HTNN ranks first among 36 methods on 200 classification datasets and third among 34 methods on 100 regression datasets. Experimental results show that the proposed hyperbolic manifold constrained tabular neural network is effective. --- ### Manifold Constrained Tabular Deep Neural Networks Authors: Tian Li, Lucy Robinson, Varun Ojha, Huizhi Liang Published: 2026-06-23 Categories: cs.LG, stat.ML arXiv: https://arxiv.org/abs/2607.09710v1 Tabular classification is often governed by local, condition-triggered rules rather than smooth global patterns. However, tabular deep neural networks (DNNs) are typically built upon Euclidean representations that favor smooth variations and semantic locality. This potential geometric mismatch can make it challenging for tabular DNNs to efficiently represent the discrete, rule-partitioned structures often underlying tabular classification. To address this issue, we propose HDE-Net, a manifold-constrained DNN that enables hierarchical decision modeling in hyperbolic space. We first abstract heterogeneous features into unified Latent Decision Nodes (LDNs) and embed them in the Poincaré ball, forming a continuous representation that resembles tree-structured reasoning. For numerical features, we introduce a Soft Decision Routing mechanism that approximates range-based local rules in a differentiable manner, bringing their LDN semantics closer to those of categorical features. An entropy-aware capacity allocation algorithm further adapts the number of LDNs per numerical feature to balance expressiveness and complexity. On the TALENT-tiny-core classification benchmark (30 datasets), HDE-Net achieves the \textit{best average rank}, outperforming both industrial GBDTs and recent tabular DNNs while maintaining high efficiency. --- ### The Verifier is the Curriculum: Precision Sets the Return on Search in Code Self-Distillation Authors: Chenyu Zhou, Qiliang Jiang, Shuning Wu, Xu Zhou Published: 2026-06-23 Categories: cs.AI, cs.SE arXiv: https://arxiv.org/abs/2607.09709v2 Post-training a code generator against a learned judge can optimize proxy features that raise the score without improving the artifact. We study the opposite signal: a deterministic, judge-free filter that asks only whether a generated project launches cleanly under a headless engine (strict-launch). Under this gate, rejection-sampling self-distillation compounds out-of-family generalization: on GameCraft-Bench a 14B model raises the per-candidate clean-launch rate on four held-out families from 8.8% to 42.2% and coverage at 32 candidates from 84% to 100%, the gold references' own ceiling, beating the supervised model on every one of the 25 held-out tasks. The gate costs one engine invocation per candidate: no reward model, no judge. At a fixed admitted count, what governs the loop is verifier precision. Swapping in a lenient build check alone erases the gain (p=0.0012); a matched gold-duplication control regresses below the supervised model. Under a semantic gate on APPS, dialing fuel precision from 1.0 to 0.25 at fixed candidate count prices that fuel linearly: over 23 training seeds, half-clean fuel returns +3.59 percentage points against the +3.69 a linear rate predicts. Under count-matched rejection-SFT only one direction of verifier error carries a measurable cost. Search obeys the same gate: quadrupling the harvest budget is worth +1.62pp behind a strict gate and nothing distinguishable from zero behind a partial-credit one. Recall is nearly free; search pays only through a precise gate: the verifier is the curriculum. --- ### The Verifier is the Curriculum: Execution-Gated Self-Distillation for Cross-Family Game Generation Authors: Chenyu Zhou, Qiliang Jiang, Shuning Wu, Xu Zhou Published: 2026-06-23 Categories: cs.AI, cs.SE arXiv: https://arxiv.org/abs/2607.09709v1 Post-training a code generator against a learned judge can optimize proxy features that raise the score without improving the artifact. We study the opposite signal: a deterministic, judge-free, ungameable filter -- whether a generated project launches cleanly under a headless engine (strict-launch). Under this gate, rejection-sampling self-distillation compounds out-of-family generalization. On GameCraft-Bench (mapping a natural-language brief to a complete Godot project), a 14B model (Qwen3-14B+LoRA) distilled under strict-launch raises clean generation on four unseen game families from 8.8% to 42.2% per-candidate and best-of-K coverage from 18/25 to 25/25 (the gold ceiling) over three rounds, each a significant gain (p=0.0019, p<1e-4, p<1e-4). The gain is not from merely adding data: an exactly-matched gold-duplication control regresses below the base model (5.6% vs. 8.8%, p=0.019), while a count-matched decomposition splits the round-1-to-2 jump into comparable quality (+8.8pp) and quantity (+8.5pp) channels. Most directly, rerunning the loop with only the filter swapped -- the lenient BUILD check, which passes 99.9% of generations, in place of the launch gate -- erases the gain entirely (back to base, p=1e-3 vs. the launch-gated round), isolating verifier precision rather than the optimizer. A second ungameable signal, headless execution grounding, rises monotonically across rounds and yields far more grounded candidates than gold-duplication at a matched budget (16 vs. 5), confirming the gains are functional, not launch-but-empty. Game generation is a verifiable testbed for one lesson: the verifier is the curriculum -- what it certifies is what the model learns. --- ### GES-TSP: Graph Edge Sparsification for TSP Authors: Tianfeng Chen, Xianyue Li Published: 2026-06-23 Categories: cs.AI, cs.LG, math.CO arXiv: https://arxiv.org/abs/2607.09708v1 Solving large-scale instances of the Traveling Salesman Problem (TSP) exactly is computationally expensive. Researchers often employ graph sparsification methods to improve computational efficiency. Traditional sparsification methods typically rely on fixed heuristics and fail to fully exploit instance-specific structural information. In this paper, we propose Graph Edge Sparsification (GES), a learning-based sparsification approach for Euclidean TSP. By incorporating geometric structural information and combinatorial optimization technology, our proposed method adaptively generates a sparsification graph for different instances, significantly reducing the graph size and accelerating the solving process. Experimental results demonstrate that our sparsification method can prune up to 95% of edges on the MATILDA dataset, while keeping the solution gap within 1% of the optimal value. Moreover, our approach exhibits strong generalization capability on the TSPLIB benchmark.In some large-scale instances, the pruning rate exceeds 99%, while the optimality gap remains below 1%. --- ### Text Distance from Nested and Hierarchical Repetitions: A Compression-Based Perspective Authors: Xiaojun Hu, Jing Wang, Jingwen Zhang, Fengyao Zhai, Xiao Xie, Hao Liao, Zengru Di, Yu Liu Published: 2026-06-23 Categories: cs.CL, cs.IT arXiv: https://arxiv.org/abs/2607.05416v1 We present a new method for structural sequence analysis grounded in Algorithmic Information Theory (AIT). At its core is the Ladderpath approach, which extracts nested and hierarchical relationships among repeated substructures in linguistic sequences -- an instantiation of AIT's principle of describing data through minimal generative programs. These structures are then used to define three distance measures: a normalized compression distance (NCD), and two alternative distances derived directly from the Ladderpath representation. Integrated with a $k$-nearest neighbor classifier, these distances achieve strong and consistent performance across in-distribution, out-of-distribution (OOD), and few-shot text classification tasks. In particular, all three methods outperform both gzip-based NCD and BERT under OOD and low-resource settings. These results demonstrate that the structured representations captured by Ladderpath preserve intrinsic properties of sequences and provide a lightweight, interpretable, and training-free alternative for text modeling. This work highlights the potential of AIT-based approaches for structural and domain-agnostic sequence understanding. --- ### Adaptive Enhanced Quantum-inspired Simulated Bifurcation Algorithm for Population State Perception Authors: Dongmei Liu, Jian Li, Xiubo Chen, JinTao Wang Published: 2026-06-23 Categories: cs.NE arXiv: https://arxiv.org/abs/2607.02540v1 Existing quantum-inspired simulated bifurcation algorithms rely on dynamic scheduling methods but lack the ability to adapt effectively to different problem instances. Additionally, during the evolutionary stage, balancing exploration and exploitation remains challenging. The fundamental issue stems from the widespread use of static preset parameters and globally uniform strategies, which can diminish algorithm effectiveness and lead to result homogenization. This article proposes an Adaptive Enhanced Quantum-inspired Simulated Bifurcation (AE-QSB) framework driven by population states. By leveraging perception indicators of four distinct population states, the QSB algorithm establishes a closed-loop strategy encompassing perception, decision-making, and execution. Within this framework, we introduce three complementary algorithms spanning a spectrum from efficient extremum seeking (ME-BSB), through population-level uniform refinement (SE-DSB), to density-aware adaptive scheduling (SG-DSB). On the medium-sized graph G22, both SE-DSB and SG-DSB achieve a mean gap below 0.05\%, while ME-BSB attains the optimal trade-off between runtime and solution quality with a gap of 0.26\% and the shortest single-run time. We compared AE-QSB variants with other algorithms across all benchmark graphs from G1 to G81. The results demonstrate that AE-QSB achieved the lowest mean gap on 74.6\% of the graphs and the highest average approximation rate on 84.5\% of the graphs. Ablation experiments further revealed that subgroup exploration and rescue mechanisms play crucial roles in both multifactor and single-factor components. This study demonstrates that population statistical information during dynamic evolution provides a computable and effective foundation for adaptive control, enabling quantum-inspired optimization methods to transition from fixed scheduling to data-driven closed-loop control. --- ### GPU-Accelerated Inverse Structural Anastylosis from Block Collapse Dynamics Authors: L. A. Muñoz Published: 2026-06-23 Categories: cs.CV, cs.LG arXiv: https://arxiv.org/abs/2606.28394v1 The physical anastylosis of collapsed architectural monuments -- the meticulous reassembly of fallen stone elements into their original structural configuration -- represents one of the most intellectually demanding challenges in conservation science. Traditional approaches depend heavily on expert archaeologist judgement and manual block-by-block correspondence, a process that is both labour-intensive and inherently subjective. Inspired by the combinatorial complexity of this problem as manifested in the game of Jenga, we present Jenga Inverse Predictor , a GPU-accelerated deep learning framework that addresses structural anastylosis as an inverse prediction task. Given an image of a collapsed block assembly, JIP-2 reconstructs the most probable prior tower configuration by: (1) implementing a complete rigid-body physics engine with OBB/SAT collision detection and a Projected Gauss-Seidel (PGS) contact solver accelerated with Numba JIT and CuPy CUDA; (2) applying the analytical force thresholds of Ziglar (CMU, 2006) -- F_app = 3*mu_s*m*g (Y-axis, torque-free) and F_app = 4*mu_s*m*g (X-axis, torque risk) -- over three friction levels (mu_s in {0.25, 0.40, 0.60}) across 450 simulated episodes; (3) training a dual-stream ResNet-18 that injects a friction one-hot vector and jointly predicts block removal count, per-position removal probabilities, centre-of-mass imbalance, and Ziglar torque risk; and (4) generating a smooth 3-D video of the block-by-block reverse reconstruction. We discuss implications for computer-assisted anastylosis at the UNESCO Maya site of Uxmal, Yucatan, and provide a detailed technical description of the full pipeline, architecture, and loss formulation. --- ### Transition-Aware best-of-N sampling for Longitudinal Chest X-ray Reports Authors: Halil Ibrahim Gulluk, Max Van Puyvelde, Wim Van Criekinge, Olivier Gevaert Published: 2026-06-23 Categories: cs.CV arXiv: https://arxiv.org/abs/2606.28393v1 In longitudinal clinical practice, every chest X-ray is read in the context of the patients prior exam, and much of what the radiologist communicates is the change from one visit to the next. To the best of our knowledge, we present the first training-free best-of-N sampling scheme for pre-trained chest X-ray report generators that is explicitly aware of this longitudinal prior to current transition. We call it transition-aware best-of-N sampling, each report is split into sentences and embedded into an unordered set in Rd; each (prior, current) pair is reduced to a fixed-dim directional vector via a set-to-set distance designed to encode the change between the two sets; and candidates are scored by cosine distance from their candidate transition vector to a cached bank of ground-truth training transition vectors, aggregated as min or kNN. We instantiate the framework with four directional set distances (mean-shift, novelty residual, directed-Hausdorff anchor, and cost-weighted optimal transport) and evaluate on a multi-visit AP-PA cohort, running inference under three prompts on three vision-language generators. Transition-aware best-of-N outperforms random selection across the board, with the largest relative gains on the Impression section. --- ### RADIANT-PET: Reasoning-Augmented PET/CT Lesion Segmentation with Large Language Models and Reinforcement Learning Authors: Jiasheng Wang, Tanun Jitwatcharakomol, Piyawadee Jongpradubgiat, Simeng Zhu Published: 2026-06-23 Categories: cs.CV, cs.AI, cs.LG arXiv: https://arxiv.org/abs/2606.28392v1 Accurate lesion segmentation in PET/CT is critical for oncology, yet remains challenging because physiologic tracer uptake and artifacts can mimic malignant signal. We present RADIANT-PET, a reasoning-augmented framework that couples a high-sensitivity voxel-level segmentation model with lesion-level large language model (LLM) adjudication. Candidate uptake regions are generated with a deliberately permissive segmentation stage, then converted into structured textual descriptions that summarize uptake intensity, morphology, and regional and global anatomical context. An LLM classifies each candidate as true lesion vs. false positive, optionally leveraging the radiology report as additional clinical context. To strengthen lesion-level reasoning, we further optimize a local LLM via reinforcement learning using Group Relative Policy Optimization, rewarding correct lesion classification and anatomically concordant site assignment. Across AutoPET and an OSU test cohort, RADIANT-PET consistently outperforms strong image-only baselines, with the largest improvements observed when radiology reports are provided. Overall, these results demonstrate that LLM-based lesion-level reasoning adds a novel reasoning layer beyond conventional segmentation, suppressing physiologic false positives and aligning voxel-level predictions with clinical interpretation. The project repository is available at: https://github.com/jwang-580/RADIANT-PET. --- ### Few-class Fidelity: Evaluating Explanations of Real-conditions CNN classifiers with Optimized Perturbations Authors: Wistan Marchadour, Pedro Soto Vega, Franck Vermet, Mathieu Hatt Published: 2026-06-23 Categories: cs.CV, cs.AI arXiv: https://arxiv.org/abs/2606.28391v2 The wide use of Convolutional Neural Networks (CNN) in numerous domains and real-world classification applications is justified by their high precision and automation speed, helping users concentrate on higher-expertise tasks. To better understand the models and avoid bias during deployment, eXplainable Artificial Intelligence (XAI) techniques can be used after training. But as the list of XAI solutions expand, comparisons between them diverge, and consensus over their evaluation cannot be reached. This paper proposes a variation of Fidelity-based XAI metrics, with a focus on real-conditions applications, where the number of classes is often low. The approach generates in-distribution, uncertainty-provoking perturbations, to ensure proper measurement of the XAI methods faithfulness. As demonstration of the evaluation framework usefulness, it is compared with human-centric object localization and segmentation metrics. Once applied to both medical and natural imaging applications, it highlights the intricate correlation between domain, data curation, and XAI solution choices in order to validate training of a new CNN model. --- ### Few-class Fidelity: Evaluating Explanations of Real-conditions CNN classifiers with Optimized Perturbations Authors: Wistan Marchadour, Pedro Soto Vega, Franck Vermet, Mathieu Hatt Published: 2026-06-23 Categories: cs.CV, cs.AI arXiv: https://arxiv.org/abs/2606.28391v1 The wide use of Convolutional Neural Networks (CNN) in numerous domains and real-world classification applications is justified by their high precision and automation speed, helping users concentrate on higher-expertise tasks. To better understand the models and avoid bias during deployment, eXplainable Artificial Intelligence (XAI) techniques can be used after training. But as the list of XAI solutions expand, comparisons between them diverge, and consensus over their evaluation cannot be reached. This paper proposes a variation of Fidelity-based XAI metrics, with a focus on real-conditions applications, where the number of classes is often low. The approach generates in-distribution, uncertainty-provoking perturbations, to ensure proper measurement of the XAI methods faithfulness. As demonstration of the evaluation framework usefulness, it is compared with human-centric object localization and segmentation metrics. Once applied to both medical and natural imaging applications, it highlights the intricate correlation between domain, data curation, and XAI solution choices in order to validate training of a new CNN model. --- ### Automated Quality Assessment of Geospatial Vector Data: A GeoAI Approach using Spatial Representation Learning Authors: Hao Li, Chen Chu, Filip Biljecki, Cyrus Shahabi, Wenwen Li Published: 2026-06-23 Categories: cs.CV, cs.AI arXiv: https://arxiv.org/abs/2606.28390v1 Geospatial vector data quality is a foundational research topic in GIS, yet classic rule-based quality assessment algorithms often struggle with diverse urban morphologies and massive data volumes. Recently, Geospatial Artificial Intelligence (GeoAI) shows promising potential for automating geospatial analysis, while its application to native vector data remains largely underexplored. To fill this research gap, we proposed Topo4Vec, an automated GeoAI framework, designed for scalable vector data quality assessment via advanced Spatial Representation Learning (SRL). Specifically, Topo4Vec relax the labor-intensive manual annotation process via topological error simulation, such as overlapping polygons and street network connectivity errors e.g., overshoots and undershoots. Then, it leverages state-of-the-art SRL approaches to encode complex, native vector geometries (e.g., polylines and polygons) into a latent space where topological errors are isolated from valid ones. A systematic performance evaluation across three study areas (Los Angeles, Munich, and Singapore) demonstrates the effectiveness and robustness of Topo4Vec, achieving a peak accuracy of 0.99 for detecting overlapping building footprints and 0.60 for overshoots and undershoots in street networks. Moreover, lessons learned from Topo4Vec shed a promising light into a scalable and autonomous GeoAI approach for large-scale vector data consistency and quality monitoring within the fast-growing geospatial data ecosystems. The code and data used in the paper are made openly available in https://figshare.com/s/612148eeb4bccadbd715. --- ### SoccerNet 2026 Player-Centric Ball Action Spotting: Per-Player Attention with Agreement-Based Ensembling Authors: Faisal Altawijri, Ismail Mathkour Published: 2026-06-23 Categories: cs.CV arXiv: https://arxiv.org/abs/2606.28389v1 We present our submission to the SoccerNet 2026 Player-Centric Ball Action Spotting challenge, which uses a two-stage pipeline: a Track-Aware Action Detector (TAAD) produces per-player action logits from broadcast video, and a Denoising Sequence Transduction (DST) transformer converts game-state features and TAAD logits into structured event sequences. We improve the TAAD with a temporal transformer that adds cross-frame context, alongside several training fixes. For the DST stage, we introduce a two-stage per-player attention mechanism operating on game-state features, and show that a spatial-first attention ordering (cross-player attention before temporal attention) improves validation Macro-F1 by 1.87%. To exploit architectural diversity, we train four model variants and combine them with a Weighted Event Fusion ensemble that applies agreement filtering to suppress single-model false positives while preserving recall, plus a dedicated exception for the rare tackle class. Our final system improves the challenge Macro-F1 from a baseline of 48.6 to 58.94. --- ### Schema-First Retrieval: Embedding Catalogs for Natural Language Analytics Authors: Adarsh Agrawal, Shashank Indukuri Published: 2026-06-23 Categories: cs.IR, cs.AI arXiv: https://arxiv.org/abs/2606.28387v1 Enterprise text-to-SQL systems often fail before SQL is generated: the model receives the wrong schema context. Modern warehouses contain thousands of tables, abbreviated columns, informal metrics, hidden join conventions, and permission boundaries that are not captured by raw table names. We introduce Schema-First Retrieval, a retrieval layer that embeds catalog metadata rather than warehouse rows. The system indexes five typed catalog objects, tables, columns, metrics, relationships, and query history, using object-specific text templates. At query time, it combines parallel vector search, lineage expansion, cross-encoder reranking, workload memory, and deterministic access-control gates before SQL generation. On CRUSH4SQL (1,534 questions), Schema-First Retrieval reaches 96.4% table recall@20 and cross-encoder reranking adds +11.1 points at column recall@10; against an equally-templated BM25 baseline, semantic retrieval is +32.8 points at table recall@5. On SEDE (857 questions), query history raises table recall@5 from 52.1% to 92.3%. On BIRD (96 questions), schema-first context reduces SQL execution errors from 15.6% to 6.2%, a 2.5x reduction. These results show that catalog selection is a first-class retrieval problem for natural language analytics, not a prompt formatting detail. --- ### Test-Input Generation for Tensor Programs: What Actually Finds Kernel Bugs Authors: Dipankar Sarkar Published: 2026-06-23 Categories: cs.SE, cs.LG arXiv: https://arxiv.org/abs/2606.27396v2 Test-input generation for tensor kernels is folkloric. Most projects pick a representative shape and dtype, run a fixed-shape allclose-style check, and ship. We make the choices explicit and measure them. Using the gpuemu op-schema-aware seeded fuzzer (arXiv:2606.20128), we evaluate seven test-generation strategies across a 26-op corpus (16 correct controls and 10 LLM-style buggy variants seeded with documented transcription patterns) on an RTX 3060 GPU instance. Strategies vary the shape candidate set, the dtype mix, and the input value distribution. We report each strategy on two axes: bug recall and control false-positive (FP) rate. Boundary-only shape sampling is the operationally safe winner: 78% recall on the 10 buggy kernels with 0% FP on the 16 controls. Adversarial value sampling reaches higher recall (99%) but inflates control FP to 94% because the strategy injects NaN and Inf inputs and the validator's NaN check fires on every kernel that propagates them, not only on buggy kernels. On the two softmax tail-mask bugs the "regular" strategy (no boundary shapes) catches 0%, while boundary raises recall to 100% and 62% respectively. That gap is the clearest single signal in the data. The corpus result is about which seeded bug patterns each strategy catches, not about the bug rate of any specific deployed LLM. --- ### Life After Benchmark Saturation: A Case Study of CORE-Bench Authors: Nitya Nadgir, Sayash Kapoor, Kangheng Liu, Peter Kirgis, Matilda Orona, Stephan Rabanser, Tilman Bayer, Abhishek Shetty, Yue Ling, Derrick Chan-Sew, Rumi Nakagawa, Saiteja Utpala et al. Published: 2026-06-23 Categories: cs.AI arXiv: https://arxiv.org/abs/2606.26158v1 When a benchmark's accuracy saturates, it is often retired and replaced with a more challenging version. We show that this approach privileges accuracy and misses the opportunity to study six other key dimensions of agent performance: construct validity issues such as shortcuts, out-of-distribution generalizability, efficiency, reliability, the relative importance of the model versus the scaffold, and uplift from human-agent collaboration. We use CORE-Bench Hard, a benchmark for computational reproducibility of scientific code, as a case study to demonstrate that measuring agents along these dimensions yields meaningful insights into agent performance even after accuracy saturates. First, we surface threats to construct validity in CORE-Bench Hard that are difficult to anticipate with less capable agents. We introduce an improved benchmark, CORE-Bench v1.1, and an out-of-distribution task suite, CORE-Bench OOD. Second, we find that despite accuracy saturation, CORE-Bench v1.1 remains useful for measuring efficiency, reliability, model performance, and scaffold performance. Finally, we conduct a small-scale randomized experiment to measure uplift from human-agent collaboration on real-world computational reproducibility tasks. We find a statistically significant speedup by about a factor of two -- likely underestimated due to one-fifth of human-only reproductions reaching the time limit before completing -- and describe various other findings. Together, our contributions present a more rigorous alternative to the dominant accuracy-centric evaluation paradigm. --- ### Reducing Redundancy in Whole-Slide Image Patching for Scalable Indexing and Retrieval Authors: Jialiang Geng, Ghazal Alabtah, Saghir Alfasly, Wataru Uegami, H. R. Tizhoosh Published: 2026-06-23 Categories: cs.IR, cs.AI arXiv: https://arxiv.org/abs/2606.26157v1 The rapid growth of digital pathology has created an urgent need for efficient indexing and retrieval of whole slide images (WSIs). This need is intensified by emerging generative AI workflows, particularly retrieval-augmented generation (RAG), which require dependable similarity search to support high-stakes clinical decision-making. Yet the substantial cost of high-performance storage limits the scalability and accessibility of WSI indexing for many healthcare institutions. Consequently, methods that can reduce storage demands while preserving retrieval accuracy have become a critical research priority. We propose ARReST (Antithetical Redundancy Reduction Strategy), a principled oppositional framework that leverages redundancy across dissimilar tissue classes to markedly decrease the number of patches that must be indexed from each WSI. Instead of eliminating only within-class duplicates, ARReST identifies antithetical patches-those whose representations contribute minimally to cross-class discrimination-and prunes them from the searchable archive. This targeted reduction substantially compresses the index without sacrificing morphological diversity or retrieval fidelity. By minimizing superfluous patch representations, ARReST reduces storage footprint, lowers computational overhead, and accelerates similarity search across large pathology repositories. Extensive experiments on TCGA repository (The Cancer Genome Atlas with 21 organs) demonstrate that ARReST achieves significant index compression while maintaining competitive retrieval performance. The observed storage savings of 3% to 60% (14%$\pm$13%) can be reliably achieved without compromising retrieval performance for many organs. The proposed strategy enables scalable, cost-efficient WSI indexing and is well-suited for next-generation retrieval-driven clinical AI systems. --- ### Detecting and Controlling Sycophancy with Cascading Linear Features Authors: Maty Bohacek, Rishub Jain, Nicholas Dufour, Thomas Leung, Chris Bregler, Roma Patel Published: 2026-06-23 Categories: cs.AI arXiv: https://arxiv.org/abs/2606.26155v1 Interpreting and controlling model behaviors through activation steering methods requires many pairs of contrastive samples that clearly exhibit desired or undesired behavior. These data pairs determine the degree to which interpretability frameworks can reliably detect model features responsible for a behavior, and therefore the ability to steer models toward or away from such behavior. In this work, we present an iterative data generation pipeline that isolates cascading linear features responsible for a behavior. Specifically, we show how moving beyond simple binary pairs of samples, and instead isolating samples that show degrees of features that scale linearly with behavior, allows for better disentanglement of features. We focus on detecting and steering away from sycophancy -- the tendency of language models to prioritize user validation. We demonstrate that sycophancy features discovered through cascading samples form linearly separable subspaces, and allow for selection of model activations that more clearly correspond to the desired behavior than baseline approaches. We also evaluate their ability to enable detection, deterministic scoring, and robust steering, and see that they either match or outperform LLM-as-a-judge and system prompting baselines while providing lower computational demand and more interpretability guarantees. Code & Data: https://cascading-feats.github.io/ --- ### Reinforcement Learning Enables Autonomous Microrobot Navigation and Intervention in Simulated Blood Capillaries Authors: Jannik Drotleff, Samuel Tovey, Paul Hohenberger, Christoph Lohrmann, Julian Hoßbach, Konstantin Nikolaou, Christian Holm Published: 2026-06-23 Categories: cs.RO, cs.LG, physics.bio-ph arXiv: https://arxiv.org/abs/2606.26154v1 Autonomous microrobots navigating biological vasculature could enable targeted drug delivery and thrombolysis, yet training control policies for realistic environments remains an open challenge. Prior reinforcement learning (RL) studies of microrobotic navigation have been limited to idealized geometries that omit complex hydrodynamic flow fields, confined branching structures, and dense cellular obstacles found in vivo. Here, we develop a physically grounded simulation of a blood capillary network, incorporating realistic hydrodynamic flow fields, explicit red blood cell dynamics, and anatomically derived branching geometry, and train deep RL agents to navigate it via chemotaxis. We systematically map the physical limits of navigation across robot size and swimming speed, revealing a forbidden regime where Brownian motion and flow overcome propulsion. Successful agents independently discover multiple universal strategy types, including run-and-rotate and energy-efficient search-and-sit policies, regardless of robot parameters. Without retraining, these agents perform targeted blocking and unblocking of capillary flow, restoring throughput to healthy baseline levels. These results establish RL as a viable framework for developing autonomous microrobotic intervention strategies in complex biological environments. --- ### Unsupervised Memory-Enhanced Video Transformers: Obstacle Detection for Autonomous Agricultural Rover Authors: Théo Biardeau, Anne-Sophie Capelle-Laizé, Salwan Alwan, David Helbert Published: 2026-06-23 Categories: cs.RO, cs.AI arXiv: https://arxiv.org/abs/2606.26151v1 While autonomous rovers have become indispensable to precision farming, achieving consistent operational safety remains a critical challenge. Conventional safety sensors, such as LiDAR, fail to detect obstacles positioned below the plant canopy, posing a significant risk. While camera-based supervised learning methods can detect common objects, they perform poorly when faced with obstacles that were not present in their training data. Actual unsupervised anomaly detection offers a solution by learning the normal visual patterns of an environment, but often fails for the dynamic scenes captured by a moving rover.\\ This paper introduces Video Memory Transformers for Anomaly Detection (VMTAD), a fully unsupervised method designed for real-time obstacle detection in dynamic agricultural scenes. VMTAD utilizes a transformer-driven architecture augmented with a dedicated memory module. This memory module leverages temporal context by processing encoded representations of preceding frames. This approach enables the system to effectively address the dynamic context caused by the robot's movement. The model is trained using only images that represent normal operation, requiring no data labels.\\ VMTAD was rigorously evaluated on the 'Grillion' agricultural rover. On a challenging rapeseed dataset, VMTAD achieved state-of-the-art performance, reaching a 0.973 detection and 0.997 segmentation Area Under the Receiver Operating Characteristic curve. A lightweight variant provides an optimal balance of high accuracy and real-time inference (14 ms), which is critical for safety, as confirmed by our analysis of the rover's total stopping distance. --- ### Extreme Meta-Classification for Large-Scale Zero-Shot Retrieval Authors: Sachin Yadav, Deepak Saini, Anirudh Buvanesh, Bhawna Paliwal, Kunal Dahiya, Siddarth Asokan, Yashoteja Prabhu, Jian Jiao, Manik Varma Published: 2026-06-23 Categories: cs.IR, cs.LG arXiv: https://arxiv.org/abs/2606.25237v1 We develop accurate and efficient solutions for large-scale retrieval tasks where novel (zero-shot) items can arrive continuously at a rapid pace. Conventional Siamese-style approaches embed both queries and items through a small encoder and retrieve the items lying closest to the query. While this approach allows efficient addition and retrieval of novel items, the small encoder lacks sufficient capacity for the necessary world knowledge in complex retrieval tasks. The extreme classification approaches have addressed this by learning a separate classifier for each item observed in the training set which significantly increases the representation capacity of the model. Such classifiers outperform Siamese approaches on observed items, but cannot be trained for novel items due to data and latency constraints. To bridge these gaps, this paper develops: (1) A new algorithmic framework, EMMETT, which efficiently synthesizes classifiers on-the-fly for novel items, by relying on the readily available classifiers for observed items; (2) A new algorithm, IRENE, which is a simple and effective instance of EMMETT that is specifically suited for large-scale deployments, and (3) A new theoretical framework for analyzing the generalization performance in large-scale zero-shot retrieval which guides our algorithm and training related design decisions. Comprehensive experiments are conducted on a wide range of retrieval tasks which demonstrate that IRENE improves the zero-shot retrieval accuracy by up to 15% points in Recall@10 when added on top of leading encoders. Additionally, on an online A/B test in a large-scale ad retrieval task in a major search engine, IRENE improved the ad click-through rate by 4.2%. Lastly, we validate our design choices through extensive ablative experiments. The source code for IRENE is available at https://aka.ms/irene. --- ### Structuring Sparsity: Block-Sparse Featurizers Capture Visual Concept Manifolds Authors: Thomas Fel, Matthew Kowal, Mozes Jacobs, Dron Hazra, Usha Bhalla, Lee Sharkey, Lucius Bushnaq, Satchel Grant, Tal Haklay, Thomas Icard, Can Rager, Michael Pearce et al. Published: 2026-06-23 Categories: cs.CV arXiv: https://arxiv.org/abs/2606.25234v1 What is the geometry of a visual percept? The most widely used protocols for decomposing neural network representations into interpretable parts treat concepts as isolated directions, yet recent work shows that concepts are often realized as geometric structures in low dimensional regions of activation space. We turn to the literature of Structured sparsity to close this gap, and show that block sparsity, which groups directions into blocks, is the prior matched to a generative model in which a representation is a sparse sum of low-dimensional manifolds: the modern, learned form of a classical idea in visual neuroscience, where a visual feature is carried by a coordinated group of neurons rather than a single tuned one. We implement three variants of block-sparse featurizers (BSFs) and, through a minimum-description-length analysis, show that all three describe activations more compactly than direction-based featurizers, with the recovered concepts typically two- to four-dimensional. We then use BSFs to (i) recontextualize prior work, showing that curve detectors in InceptionV1 actually read from a single continuous curve manifold, (ii) discover novel manifolds including shadows and lighting in DINOv3, and (iii) support interpretable control of image generation in diffusion models (SDXL) via manifold steering. --- ### Semantic Allocation in Ordered Bottlenecks: Predictive Residual Inference for Visual Representation Learning Authors: Erik Ayari, Manuel Traub, Martin V. Butz Published: 2026-06-23 Categories: cs.LG, cs.CV arXiv: https://arxiv.org/abs/2606.25232v1 Ordered bottlenecks aim to provide utility at flexible budgets by assigning coarse information to early tokens and task-relevant detail to later ones. Prior work, including tail dropping (TD), typically enforces ordering by means of a masking-based ordering pressure (MBOP): Late tokens are masked more frequently than early tokens and are therefore encouraged to store less essential fine details. We introduce predictive residual inference for ordered representations (PRIOR), a framework designed to address inherent weaknesses of MBOP. MBOP is prone to weak late-token utility because it lacks an explicit refinement objective and uses gradient exposure as a proxy for importance. Furthermore, representations may become particularly brittle in optimization-sensitive settings, such as when using discrete or quantized token representations. PRIOR replaces activation-rate control with log2-scaled levels and level-wise predictors. These predictors separate already explained from unexplained information, focusing each level on residual error. We compare PRIOR against MBOP-TD and independent tail-biased dropout (MBOP-ITD) in contrastive learning and image reconstruction tasks. Unlike the baselines, PRIOR learns well-ordered representations across experiments: low budgets provide coarse descriptors, while high budgets add refinements. Simultaneously, full-budget performance with PRIOR is higher in all but one experimental setting, where performance remains comparable. MBOP baselines are severely limited in discrete and quantized settings, while PRIOR approaches the performance of continuous counterparts. Taken together, these findings establish PRIOR as an effective framework for ordered representation learning. --- ### Towards Structuring an Arabic-English Machine-Readable Dictionary Using Parsing Expression Grammars Authors: Diaa M. Fayed, Aly A. Fahmy, Mohsen A. Rashwan, Wafaa K. Fayed Published: 2026-06-23 Categories: cs.CL arXiv: https://arxiv.org/abs/2606.25231v3 Dictionaries are rich sources of lexical information about words that is required for many applications of natural language processing and human language technology. However, publishers prepare printed dictionaries for human usage not for machine processing. This paper presented a method to structure partly a machine-readable version of the Arabic-English Al-Mawrid dictionary. The method converted the entries of Al-Mawrid from a stream of words and punctuation marks into hierarchical structures. The hierarchical structure expresses the components of each dictionary entry in explicit format. A dictionary entry is composed of subentries and each subentry consists of defining phrases, domain labels, cross-references, and translation equivalences. We designed the proposed method as cascaded steps where parsing is the main step. We implemented the parser using the parsing expression grammars formalism. In conclusion, although Arabic dictionaries do not have microstructure standardization, this study demonstrated that it is possible to structure them automatically or semi-automatically with plausible accuracy after inducing their microstructure. --- ### MJEPA: A Simple and Scalable Joint-Embedding Predictive Architecture for Audio-Visual Learning Authors: Revant Teotia, Adrien Bardes, Michael Rabbat, Sumit Chopra, Matthew J. Muckley, Nicolas Ballas Published: 2026-06-23 Categories: cs.CV, cs.LG arXiv: https://arxiv.org/abs/2606.25225v1 Self-supervised learning from large-scale video data has emerged as a dominant paradigm for visual representation learning. Since audio and visual streams naturally co-occur in video data, extending this success to jointly learn from both modalities is a natural next step, yet it remains challenging. Existing audio-visual self-supervised methods rely on modality-specific encoders and complex combinations of contrastive or reconstruction objectives, limiting cross-modal synergy and scalability. Joint Embedding Predictive Architectures (JEPAs) offer a simple, modality-agnostic alternative, but have to date been applied primarily to individual modalities. We introduce MJEPA, a joint-embedding predictive architecture for audio-visual learning that uses a single, unified encoder for both modalities. Our approach uses only a single predictive objective, applied both within and across modalities. We show that cross-modal prediction is critical: without it, a shared encoder degrades below unimodal baselines; with it, each modality's representation benefits from the other. Our frozen ViT-g model outperforms the best prior frozen baseline by over 6.8 mAP on AudioSet-20K, surpasses fully finetuned models on ESC-50 and FSD50K, and is competitive on video benchmarks despite using 10x less video data. --- ### Cage-based Texture Transfer with Geometric Filtering Authors: Rose Mei Zhou, Lynnette Hui Xian Ng, Adrian Xuan Wei Lim, Conor Griffin, Faraz Baghernezhad Published: 2026-06-23 Categories: cs.CV, cs.GR arXiv: https://arxiv.org/abs/2606.25220v1 Real-time texture transfer expands the creative horizon for interactive applications, enabling seamless detail projection in scenarios that range from digital character cosmetics to procedural automotive texturing. Yet, its practical application is governed by inherent trade-offs between processing speed and suppression of artifacts. Low-latency transfer methods frequently fail to suppress artifacts, and robust alternatives rely on large-scale models that are costly in training and memory. Our proposed method bridges the gap between efficiency and robustness by using a cage-based geometric filtering method to identify Non-Cosmetic Zones (NCZs) for artifact suppression. While other models are resource-intensive and require multiple days of training on manually annotated datasets, we are able to successfully suppress artifacts and achieve immediate deployment on consumer-grade hardware. Our framework achieved highly efficient runtimes of ~70ms on mobile devices for a ~4.8k triangle mesh. --- ### Homomorphic Encryptions for Privacy Preserving Vision Authors: Preey Shah, Rohan Virani, Sanjari Srivastava Published: 2026-06-23 Categories: cs.CR, cs.CV arXiv: https://arxiv.org/abs/2606.25216v1 Legal requirements might prevent organizations from sharing sensitive data like medical or financial details of consumers which prevents them from leveraging cloud based ML-as-a-service solutions provided by third party providers, which are quickly gaining popularity these days. In this project, we aim to perform inference tasks in Computer Vision in a privacy-preserving manner, i.e, by only looking at encrypted data. Recent advances in fully homomorphic encryption make this possible. A fully homomorphic encryption allows an arbitrary sequence of additive and multiplicative operations to be performed on encrypted data directly. Applying homomorphic encryptions to CNNs requires modifying the conventional CNN layers, so that they adhere to the encryption scheme. Our aim was to explore the best methods to create CNNs which can classify encrypted images directly. We used Microsoft SEAL for performing homomorphic encryption. The performance of these "encryption based CNNs" should be comparable with baseline accuracies of the same CNNs trained on unencrypted data, and the aim was to achieve as low of a hit on inference-time performance as possible. We successfully obtained minimal drop in classification accuracy for various datasets. We used MNIST as our baseline, which is popularly used in related research work and then explored more complex datasets like Kuzushiji MNIST, Fashion-MNIST and CIFAR-10 as a part of our contribution. Additionally, we also added support for more complex operations on top of TenSEAL, like processing colored images (multi-channel input), applying multiple convolutional layers and performing average pooling. --- ### Reflective VLA: In-Context Action Consequences Make VLAs Generalize Authors: Qing Lian, Kent Yu, Lei Zhang Published: 2026-06-23 Categories: cs.CV, cs.RO arXiv: https://arxiv.org/abs/2606.25215v1 Most vision-language-action (VLA) models are reactive: they predict the next action from the current instruction and observation, implicitly assuming that the current observation fully specifies the action-relevant state. In embodied control, however, embodiment-specific factors such as camera-to-robot geometry, robot calibration, or systematic actuation bias are often hard to identify from a single observation. As a result, reactive policies cannot reliably disambiguate these factors in general, overfitting to training environments and generalizing poorly at deployment. We propose Reflective VLA, which conditions each decision on a context of observation-action-consequence triplets. Each triplet records not only what the robot observed and executed, but also how the scene changed afterward, exposing the deployment-specific mapping from actions to observed effects. Architecturally, Reflective VLA routes all observation modalities through the VLM under shared attention, so the action expert reasons directly over past triplets and the current observation. A block-causal mask enables parallel multi-frame training without leakage and supports KV-cached real-time inference. On standard LIBERO and SimplerEnv-Bridge, Reflective VLA preserves strong in-distribution performance. Under distribution shift on LIBERO-Plus and the harder LIBERO-Plus-Hard, it improves average success rate by 5.4 and 4.2 percentage points over a matched reactive baseline. Ablations with a matched history-only baseline further show that action consequences -- rather than additional context length alone -- are the key to cross-environment generalization. Project page: https://lianqing11.github.io/reflective-vla-page/ --- ### ASAP: Agent-System Co-Design for Wall-Clock-Centered Auto HPO Research for ML Experiments Authors: Taicheng Guo, Haomin Zhuang, Kehan Guo, Yujun Zhou, Nitesh V. Chawla, Olaf Wiest, Xiangliang Zhang Published: 2026-06-23 Categories: cs.LG, cs.AI, cs.CL arXiv: https://arxiv.org/abs/2606.25207v1 Hyperparameter Optimization (HPO) is essential for maximizing machine learning model performance, and its core challenge is sample efficiency: finding strong configurations within a limited budget. Because every HPO tool relies on a surrogate prior that imparts its own inductive bias, individual tools struggle once problems become sufficiently diverse and drift from these priors. Motivated by the reasoning and generalization capabilities of LLMs, recent work has explored using LLMs for HPO and reports improved per-iteration performance. Yet these methods share two limitations with a common origin: they use the LLM as a single-tool replacement evaluated by iteration count. (i) Deployed in place of prior tools, the LLM is itself constrained by its pretraining objective to one family of inductive-biased proposals; this single-source setup still fails to handle the full diversity of problems. (ii) Per-iteration evaluation ignores that, in real runs, LLM inference or tool execution is paid serially on top of model evaluation every round, so iteration-count gains do not translate into end-to-end wall-clock gains. We present ASAP, an agent-system co-design that addresses both limitations. On the agent side, ASAP uses the LLM to integrate a diverse pool of inductive-biased optimizers and to select among their proposals each round. On the system side, ASAP re-architects the loop to reduce end-to-end wall-clock while preserving regret quality: a prefix-stable prompt maximizes KV-cache reuse across rounds; speculation parallelism hides the remaining LLM and tool latency under model evaluation via a relative-error accept test; and a Self-Tuner adapts the speculation threshold from execution logs off the critical path. Extensive experiments on diverse modern HPO tasks show that ASAP consistently outperforms baselines, underscoring the value of tool integration and agent-system co-design. --- ### RAVEN: Long-Horizon Reasoning & Navigation with a Visuo-Spatio-Temporal Memory Authors: Yixun Hu, Zhicheng Zheng, Lihan Zha, Chunwei Xing, Rajdeep Singh, Omar Hossain, Antonio Loquercio, Dhruv Shah Published: 2026-06-23 Categories: cs.RO, cs.AI, cs.CL arXiv: https://arxiv.org/abs/2606.25206v1 Long-term robot deployment requires a compact and scalable memory that preserves fine-grained visual semantics, grounds observations in space and time, and enables efficient storage and retrieval. In this paper, we propose RAVEN, an agentic memory system for long-horizon robotic question answering and navigation. RAVEN stores visual embeddings with pose and time in a vector database, and grounds retrieval in a spatial map to answer queries and navigate to goals. By operating directly on visual embeddings, RAVEN avoids lossy image-to-text captioning and enables accurate semantic, spatial, and temporal retrieval at scale. Across several simulated and real-world video question-answering benchmarks, RAVEN consistently surpasses caption-based memory systems and matches frontier VLMs on long-horizon tasks at 10$\times$ lower retrieval cost. Finally, we instantiate RAVEN on a Unitree Go1 robot for the task of long-horizon navigation for natural language goal-reaching, and show successful deployment over several large indoor environments. --- ### FDN: Interpretable Spatiotemporal Forecasting with Future Decomposition Networks Authors: Nicholas Majeske, Ariful Azad Published: 2026-06-23 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2606.25201v1 Spatiotemporal systems comprise a collection of spatially distributed yet interdependent entities each generating unique dynamic signals. Highly sophisticated methods have been proposed in recent years delivering state-of-the-art (SOTA) forecasts but few have focused on interpretability. To address this, we propose the Future Decomposition Network (FDN), a novel forecast model capable of (a) providing interpretable predictions through classification (b) revealing latent activity patterns in the target time-series and (c) delivering forecasts competitive with SOTA methods at a fraction of their memory and runtime cost. We conduct comprehensive analyses on FDN for multiple datasets from hydrologic, traffic, and energy systems, demonstrating its improved accuracy and interpretability. ---