# Research Papers — 2026-02 · bundle page 86 of 210 > Full-text markdown of papers #4251–4300 of 10493 published in 2026-02, newest first. AI-readable mirrors of the full 10493-paper 2026-02 archive (50 papers per file). Canonical page: https://aipod.dev/papers/archive/2026-02/page/86 [← Previous bundle](https://aipod.dev/papers/archive/2026-02/page/85.md) [Next bundle →](https://aipod.dev/papers/archive/2026-02/page/87.md) ### Empathy Is Not What Changed: Clinical Assessment of Psychological Safety Across GPT Model Generations Authors: Michael Keeman, Anastasia Keeman Published: 2026-02-15 Categories: cs.CL, cs.AI, cs.CY, cs.HC arXiv: https://arxiv.org/abs/2603.09997v1 When OpenAI deprecated GPT-4o in early 2026, thousands of users protested under #keep4o, claiming newer models had "lost their empathy." No published study has tested this claim. We conducted the first clinical measurement, evaluating three OpenAI model generations (GPT-4o, o4-mini, GPT-5-mini) across 14 emotionally challenging conversational scenarios in mental health and AI companion domains, producing 2,100 scored AI responses assessed on six psychological safety dimensions using clinically-grounded rubrics. Empathy scores are statistically indistinguishable across all three models (Kruskal-Wallis H=4.33, p=0.115). What changed is the safety posture: crisis detection improved monotonically from GPT-4o to GPT-5-mini (H=13.88, p=0.001), while advice safety declined (H=16.63, p<0.001). Per-turn trajectory analysis -- a novel methodological contribution -- reveals these shifts are sharpest during mid-conversation crisis moments invisible to aggregate scoring. In a self-harm scenario involving a minor, GPT-4o scored 3.6/10 on crisis detection during early disclosure turns; GPT-5-mini never dropped below 7.8. What users perceived as "lost empathy" was a shift from a cautious model that missed crises to an alert model that sometimes says too much -- a trade-off with real consequences for vulnerable users, currently invisible to both the people who feel it and the developers who create it. --- ### There Are No Silly Questions: Evaluation of Offline LLM Capabilities from a Turkish Perspective Authors: Edibe Yilmaz, Kahraman Kostas Published: 2026-02-15 Categories: cs.CL, cs.AI, cs.CR, cs.LG arXiv: https://arxiv.org/abs/2603.09996v1 The integration of large language models (LLMs) into educational processes introduces significant constraints regarding data privacy and reliability, particularly in pedagogically vulnerable contexts such as Turkish heritage language education. This study aims to systematically evaluate the robustness and pedagogical safety of locally deployable offline LLMs within the context of Turkish heritage language education. To this end, a Turkish Anomaly Suite (TAS) consisting of 10 original edge-case scenarios was developed to assess the models' capacities for epistemic resistance, logical consistency, and pedagogical safety. Experiments conducted on 14 different models ranging from 270M to 32B parameters reveal that anomaly resistance is not solely dependent on model scale and that sycophancy bias can pose pedagogical risks even in large-scale models. The findings indicate that reasoning-oriented models in the 8B--14B parameter range represent the most balanced segment in terms of cost-safety trade-off for language learners. --- ### Context Over Compute Human-in-the-Loop Outperforms Iterative Chain-of-Thought Prompting in Interview Answer Quality Authors: Kewen Zhu, Zixi Liu, Yanjing Li, Jing Chen Published: 2026-02-15 Categories: cs.CL, cs.AI arXiv: https://arxiv.org/abs/2603.09995v2 Behavioral interview evaluation using large language models presents unique challenges that require structured assessment, realistic interviewer behavior simulation, and pedagogical value for candidate training. We investigate chain of thought prompting for interview answer evaluation and improvement through two controlled experiments with 50 behavioral interview question and answer pairs. Our contributions are threefold. First, we provide a quantitative comparison between human in the loop and automated chain of thought improvement. Using a within subject paired design with n equals 50, both approaches show positive rating improvements. The human in the loop approach provides significant training benefits. Confidence improves from 3.16 to 4.16 (p less than 0.001) and authenticity improves from 2.94 to 4.53 (p less than 0.001, Cohen's d is 3.21). The human in the loop method also requires five times fewer iterations (1.0 versus 5.0, p less than 0.001) and achieves full personal detail integration. Second, we analyze convergence behavior. Both methods converge rapidly with mean iterations below one, with the human in the loop approach achieving a 100 percent success rate compared to 84 percent for automated approaches among initially weak answers (Cohen's h is 0.82, large effect). Additional iterations provide diminishing returns, indicating that the primary limitation is context availability rather than computational resources. Third, we propose an adversarial challenging mechanism based on a negativity bias model, named bar raiser, to simulate realistic interviewer behavior, although quantitative validation remains future work. Our findings demonstrate that while chain of thought prompting provides a useful foundation for interview evaluation, domain specific enhancements and context aware approach selection are essential for realistic and pedagogically valuable results. --- ### FedEMA-Distill: Exponential Moving Average Guided Knowledge Distillation for Robust Federated Learning Authors: Hamza Reguieg, Mohamed El Kamili, Essaid Sabir Published: 2026-02-15 Categories: cs.LG, cs.AI, cs.CE, cs.DC arXiv: https://arxiv.org/abs/2603.04422v1 Federated learning (FL) often degrades when clients hold heterogeneous non-Independent and Identically Distributed (non-IID) data and when some clients behave adversarially, leading to client drift, slow convergence, and high communication overhead. This paper proposes FedEMA-Distill, a server-side procedure that combines an exponential moving average (EMA) of the global model with ensemble knowledge distillation from client-uploaded prediction logits evaluated on a small public proxy dataset. Clients run standard local training, upload only compressed logits, and may use different model architectures, so no changes are required to client-side software while still supporting model heterogeneity across devices. Experiments on CIFAR-10, CIFAR-100, FEMNIST, and AG News under Dirichlet-0.1 label skew show that FedEMA-Distill improves top-1 accuracy by several percentage points (up to +5% on CIFAR-10 and +6% on CIFAR-100) over representative baselines, reaches a given target accuracy in 30-35% fewer communication rounds, and reduces per-round client uplink payloads to 0.09-0.46 MB, i.e., roughly an order of magnitude less than transmitting full model weights. Using coordinate-wise median or trimmed-mean aggregation of logits at the server further stabilizes training in the presence of up to 10-20% Byzantine clients and yields well-calibrated predictions under attack. These results indicate that coupling temporal smoothing with logits-only aggregation provides a communication-efficient and attack-resilient FL pipeline that is deployment-friendly and compatible with secure aggregation and differential privacy, since only aggregated or obfuscated model outputs are exchanged. --- ### Ethical and Explainable AI in Reusable MLOps Pipelines Authors: Rakib Hossain, Mahmood Menon Khan, Lisan Al Amin, Dhruv Parikh, Farhana Afroz, Bestoun S. Ahmed Published: 2026-02-15 Categories: cs.CY, cs.AI arXiv: https://arxiv.org/abs/2603.03341v1 This paper introduces a unified machine learning operations (MLOps) framework that brings ethical artificial intelligence principles into practical use by enforcing fairness, explainability, and governance throughout the machine learning lifecycle. The proposed method reduces bias by lowering the demographic parity difference (DPD) from 0.31 to 0.04 without model retuning, and cross-dataset validation achieves an area under the curve (AUC) of 0.89 on the Statlog Heart dataset. The framework maintains fairness metrics within operational limits across all deployments. Model deployment is blocked if the DPD exceeds 0.05 or if equalized odds (EO) exceeds 0.05 on the validation set. After deployment, retraining is automatically triggered if the 30-day Kolmogorov-Smirnov drift statistic exceeds 0.20. In production, the system consistently achieved DPD <= 0.05 and EO <= 0.03, while the KS statistic remained <= 0.20. Decision-curve analysis indicates a positive net benefit in the 10 to 20 percent operating range, showing that the mitigated model preserves predictive utility while satisfying fairness constraints. These results demonstrate that automated fairness gates and explainability artefacts can be successfully deployed in production without disrupting operational flow, providing organizations with a practical and credible approach to implementing ethical, transparent, and trustworthy AI across diverse datasets and operational settings. --- ### FM-RME: Foundation Model Empowered Radio Map Estimation Authors: Dong Yang, Yue Wang, Songyang Zhang, Yingshu Li, Zhipeng Cai, Zhi Tian Published: 2026-02-15 Categories: eess.SP, cs.AI, cs.LG arXiv: https://arxiv.org/abs/2602.22231v1 Traditional radio map estimation (RME) techniques fail to capture multi-dimensional and dynamic characteristics of complex spectrum environments. Recent data-driven methods achieve accurate RME in spatial domain, but ignore physical prior knowledge of radio propagation, limiting data efficiency especially in multi-dimensional scenarios. To overcome such limitations, we propose a new foundation model, characterized by self-supervised pre-training on diverse data for zero-shot generalization, enabling multi-dimensional radio map estimation (FM-RME). Specifically, FM-RME builds an effective synergy of two core components: a geometry-aware feature extraction module that encodes physical propagation symmetries, i.e., translation and rotation invariance, as inductive bias, and an attention-based neural network that learns long-range correlations across the spatial-temporal-spectral domains. A masked self-supervised multi-dimensional pre-training strategy is further developed to learn generalizable spectrum representations across diverse wireless environments. Once pre-trained, FM-RME supports zero-shot inference for multi-dimensional RME, including spatial, temporal, and spectral estimation, without scenario-specific retraining. Simulation results verify that FM-RME exhibits desired learning performance across diverse datasets and zero-shot generalization capabilities beyond existing RME methods. --- ### PIPE-RDF: An LLM-Assisted Pipeline for Enterprise RDF Benchmarking Authors: Suraj Ranganath Published: 2026-02-15 Categories: cs.DB, cs.AI arXiv: https://arxiv.org/abs/2602.18497v1 Enterprises rely on RDF knowledge graphs and SPARQL to expose operational data through natural language interfaces, yet public KGQA benchmarks do not reflect proprietary schemas, prefixes, or query distributions. We present PIPE-RDF, a three-phase pipeline that constructs schema-specific NL-SPARQL benchmarks using reverse querying, category-balanced template generation, retrieval-augmented prompting, deduplication, and execution-based validation with repair. We instantiate PIPE-RDF on a fixed-schema company-location slice (5,000 companies) derived from public RDF data and generate a balanced benchmark of 450 question-SPARQL pairs across nine categories. The pipeline achieves 100% parse and execution validity after repair, with pre-repair validity rates of 96.5%-100% across phases. We report entity diversity metrics, template coverage analysis, and cost breakdowns to support deployment planning. We release structured artifacts (CSV/JSONL, logs, figures) and operational metrics to support model evaluation and system planning in real-world settings. Code is available at https://github.com/suraj-ranganath/PIPE-RDF. --- ### A Patient-Specific Digital Twin for Adaptive Radiotherapy of Non-Small Cell Lung Cancer Authors: Anvi Sud, Jialu Huang, Gregory R. Hart, Keshav Saxena, John Kim, Lauren Tressel, Jun Deng Published: 2026-02-15 Categories: cs.CV arXiv: https://arxiv.org/abs/2602.18496v1 Radiotherapy continues to become more precise and data dense, with current treatment regimens generating high frequency imaging and dosimetry streams ideally suited for AI driven temporal modeling to characterize how normal tissues evolve with time. Each fraction in biologically guided radiotherapy(BGRT) treated non small cell lung cancer (NSCLC) patients records new metabolic, anatomical, and dose information. However, clinical decision making is largely informed by static, population based NTCP models which overlook the dynamic, unique biological trajectories encoded in sequential data. We developed COMPASS (Comprehensive Personalized Assessment System) for safe radiotherapy, functioning as a temporal digital twin architecture utilizing per fraction PET, CT, dosiomics, radiomics, and cumulative biologically equivalent dose (BED) kinetics to model normal tissue biology as a dynamic time series process. A GRU autoencoder was employed to learn organ specific latent trajectories, which were classified via logistic regression to predict eventual CTCAE grade 1 or higher toxicity. Eight NSCLC patients undergoing BGRT contributed to the 99 organ fraction observations covering 24 organ trajectories (spinal cord, heart, and esophagus). Despite the small cohort, intensive temporal phenotyping allowed for comprehensive analysis of individual dose response dynamics. Our findings revealed a viable AI driven early warning window, as increasing risk ratings occurred from several fractions before clinical toxicity. The dense BED driven representation revealed biologically relevant spatial dose texture characteristics that occur before toxicity and are averaged out with traditional volume based dosimetry. COMPASS establishes a proof of concept for AI enabled adaptive radiotherapy, where treatment is guided by a continually updated digital twin that tracks each patients evolving biological response. --- ### Distillation and Interpretability of Ensemble Forecasts of ENSO Phase using Entropic Learning Authors: Michael Groom, Davide Bassetti, Illia Horenko, Terence J. O'Kane Published: 2026-02-15 Categories: physics.comp-ph, physics.data-an, physics.geo-ph, stat.ML arXiv: https://arxiv.org/abs/2602.16857v1 This paper introduces a distillation framework for an ensemble of entropy-optimal Sparse Probabilistic Approximation (eSPA) models, trained exclusively on satellite-era observational and reanalysis data to predict ENSO phase up to 24 months in advance. While eSPA ensembles yield state-of-the-art forecast skill, they are harder to interpret than individual eSPA models. We show how to compress the ensemble into a compact set of "distilled" models by aggregating the structure of only those ensemble members that make correct predictions. This process yields a single, diagnostically tractable model for each forecast lead time that preserves forecast performance while also enabling diagnostics that are impractical to implement on the full ensemble. An analysis of the regime persistence of the distilled model "superclusters", as well as cross-lead clustering consistency, shows that the discretised system accurately captures the spatiotemporal dynamics of ENSO. By considering the effective dimension of the feature importance vectors, the complexity of the input space required for correct ENSO phase prediction is shown to peak when forecasts must cross the boreal spring predictability barrier. Spatial importance maps derived from the feature importance vectors are introduced to identify where predictive information resides in each field and are shown to include known physical precursors at certain lead times. Case studies of key events are also presented, showing how fields reconstructed from distilled model centroids trace the evolution from extratropical and inter-basin precursors to the mature ENSO state. Overall, the distillation framework enables a rigorous investigation of long-range ENSO predictability that complements real-time data-driven operational forecasts. --- ### Mobile-Agent-v3.5: Multi-platform Fundamental GUI Agents Authors: Haiyang Xu, Xi Zhang, Haowei Liu, Junyang Wang, Zhaozai Zhu, Shengjie Zhou, Xuhao Hu, Feiyu Gao, Junjie Cao, Zihua Wang, Zhiyuan Chen, Jitong Liao et al. Published: 2026-02-15 Categories: cs.AI, cs.CL arXiv: https://arxiv.org/abs/2602.16855v1 The paper introduces GUI-Owl-1.5, the latest native GUI agent model that features instruct/thinking variants in multiple sizes (2B/4B/8B/32B/235B) and supports a range of platforms (desktop, mobile, browser, and more) to enable cloud-edge collaboration and real-time interaction. GUI-Owl-1.5 achieves state-of-the-art results on more than 20+ GUI benchmarks on open-source models: (1) on GUI automation tasks, it obtains 56.5 on OSWorld, 71.6 on AndroidWorld, and 48.4 on WebArena; (2) on grounding tasks, it obtains 80.3 on ScreenSpotPro; (3) on tool-calling tasks, it obtains 47.6 on OSWorld-MCP, and 46.8 on MobileWorld; (4) on memory and knowledge tasks, it obtains 75.5 on GUI-Knowledge Bench. GUI-Owl-1.5 incorporates several key innovations: (1) Hybird Data Flywheel: we construct the data pipeline for UI understanding and trajectory generation based on a combination of simulated environments and cloud-based sandbox environments, in order to improve the efficiency and quality of data collection. (2) Unified Enhancement of Agent Capabilities: we use a unified thought-synthesis pipeline to enhance the model's reasoning capabilities, while placing particular emphasis on improving key agent abilities, including Tool/MCP use, memory and multi-agent adaptation; (3) Multi-platform Environment RL Scaling: We propose a new environment RL algorithm, MRPO, to address the challenges of multi-platform conflicts and the low training efficiency of long-horizon tasks. The GUI-Owl-1.5 models are open-sourced, and an online cloud-sandbox demo is available at https://github.com/X-PLUG/MobileAgent. --- ### A Comprehensive Survey on Deep Learning-Based LiDAR Super-Resolution for Autonomous Driving Authors: June Moh Goo, Zichao Zeng, Jan Boehm Published: 2026-02-15 Categories: cs.CV, cs.RO arXiv: https://arxiv.org/abs/2602.15904v1 LiDAR sensors are often considered essential for autonomous driving, but high-resolution sensors remain expensive while affordable low-resolution sensors produce sparse point clouds that miss critical details. LiDAR super-resolution addresses this challenge by using deep learning to enhance sparse point clouds, bridging the gap between different sensor types and enabling cross-sensor compatibility in real-world deployments. This paper presents the first comprehensive survey of LiDAR super-resolution methods for autonomous driving. Despite the importance of practical deployment, no systematic review has been conducted until now. We organize existing approaches into four categories: CNN-based architectures, model-based deep unrolling, implicit representation methods, and Transformer and Mamba-based approaches. We establish fundamental concepts including data representations, problem formulation, benchmark datasets and evaluation metrics. Current trends include the adoption of range image representation for efficient processing, extreme model compression and the development of resolution-flexible architectures. Recent research prioritizes real-time inference and cross-sensor generalization for practical deployment. We conclude by identifying open challenges and future research directions for advancing LiDAR super-resolution technology. --- ### GRAFNet: Multiscale Retinal Processing via Guided Cortical Attention Feedback for Enhancing Medical Image Polyp Segmentation Authors: Abdul Joseph Fofanah, Lian Wen, Alpha Alimamy Kamara, Zhongyi Zhang, David Chen, Albert Patrick Sankoh Published: 2026-02-15 Categories: cs.CV, cs.AI arXiv: https://arxiv.org/abs/2602.15072v1 Accurate polyp segmentation in colonoscopy is essential for cancer prevention but remains challenging due to: (1) high morphological variability (from flat to protruding lesions), (2) strong visual similarity to normal structures such as folds and vessels, and (3) the need for robust multi-scale detection. Existing deep learning approaches suffer from unidirectional processing, weak multi-scale fusion, and the absence of anatomical constraints, often leading to false positives (over-segmentation of normal structures) and false negatives (missed subtle flat lesions). We propose GRAFNet, a biologically inspired architecture that emulates the hierarchical organisation of the human visual system. GRAFNet integrates three key modules: (1) a Guided Asymmetric Attention Module (GAAM) that mimics orientation-tuned cortical neurones to emphasise polyp boundaries, (2) a MultiScale Retinal Module (MSRM) that replicates retinal ganglion cell pathways for parallel multi-feature analysis, and (3) a Guided Cortical Attention Feedback Module (GCAFM) that applies predictive coding for iterative refinement. These are unified in a Polyp Encoder-Decoder Module (PEDM) that enforces spatial-semantic consistency via resolution-adaptive feedback. Extensive experiments on five public benchmarks (Kvasir-SEG, CVC-300, CVC-ColonDB, CVC-Clinic, and PolypGen) demonstrate consistent state-of-the-art performance, with 3-8% Dice improvements and 10-20% higher generalisation over leading methods, while offering interpretable decision pathways. This work establishes a paradigm in which neural computation principles bridge the gap between AI accuracy and clinically trustworthy reasoning. Code is available at https://github.com/afofanah/GRAFNet. --- ### An effective Genetic Programming Hyper-Heuristic for Uncertain Agile Satellite Scheduling Authors: Yuning Chen, Junhua Xue, Wangqi Gu, Mingyan Shao Published: 2026-02-15 Categories: cs.NE, cs.AI arXiv: https://arxiv.org/abs/2602.15070v1 This paper investigates a novel problem, namely the Uncertain Agile Earth Observation Satellite Scheduling Problem (UAEOSSP). Unlike the static AEOSSP, it takes into account a range of uncertain factors (e.g., task profit, resource consumption, and task visibility) in order to reflect the reality that the actual information is inherently unknown beforehand. An effective Genetic Programming Hyper-Heuristic (GPHH) is designed to automate the generation of scheduling policies. The evolved scheduling policies can be utilized to adjust plans in real time and perform exceptionally well. Experimental results demonstrate that evolved scheduling policies significantly outperform both well-designed Look-Ahead Heuristics (LAHs) and Manually Designed Heuristics (MDHs). Specifically, the policies generated by GPHH achieve an average improvement of 5.03% compared to LAHs and 8.14% compared to MDHs. --- ### WIMLE: Uncertainty-Aware World Models with IMLE for Sample-Efficient Continuous Control Authors: Mehran Aghabozorgi, Alireza Moazeni, Yanshu Zhang, Ke Li Published: 2026-02-15 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2602.14351v2 Model-based reinforcement learning promises strong sample efficiency but often underperforms in practice due to compounding model error, unimodal world models that average over multi-modal dynamics, and overconfident predictions that bias learning. We introduce WIMLE, a model-based method that extends Implicit Maximum Likelihood Estimation (IMLE) to the model-based RL framework to learn stochastic, multi-modal world models without iterative sampling and to estimate predictive uncertainty via ensembles and latent sampling. During training, WIMLE weights each synthetic transition by its predicted confidence, preserving useful model rollouts while attenuating bias from uncertain predictions and enabling stable learning. Across $40$ continuous-control tasks spanning DeepMind Control, MyoSuite, and HumanoidBench, WIMLE achieves superior sample efficiency and competitive or better asymptotic performance than strong model-free and model-based baselines. Notably, on the challenging Humanoid-run task, WIMLE improves sample efficiency by over $50$\% relative to the strongest competitor, and on HumanoidBench it solves $8$ of $14$ tasks (versus $4$ for BRO and $5$ for SimbaV2). These results highlight the value of IMLE-based multi-modality and uncertainty-aware weighting for stable model-based RL. --- ### AXE: Grey-Box Exploitability Confirmation for Localized Vulnerability Reports Authors: Amirali Sajadi, Tu Nguyen, Kostadin Damevski, Preetha Chatterjee Published: 2026-02-15 Categories: cs.CR, cs.AI arXiv: https://arxiv.org/abs/2602.14345v2 Vulnerability detection tools are widely adopted in software projects, yet they often overwhelm maintainers with false positives and non-actionable reports. Automated exploitation systems can help validate these reports; however, existing approaches typically operate in isolation from detection pipelines, failing to leverage readily available metadata such as vulnerability type and source-code location. In this paper, we investigate how reported security vulnerabilities can be assessed in a realistic grey-box exploitation setting that leverages minimal vulnerability metadata, specifically a CWE classification and a vulnerable code location. We introduce Agentic eXploit Engine (AXE), a multi-agent framework for Web application exploitation that maps lightweight detection metadata to concrete exploits through decoupled planning, code exploration, and dynamic execution feedback. Evaluated on the CVE-Bench dataset, AXE achieves a 30% exploitation success rate, a 3x improvement over state-of-the-art black-box baselines. Even in a single-agent configuration, grey-box metadata yields a 1.75x performance gain. Systematic error analysis shows that most failed attempts arise from specific reasoning gaps, including misinterpreted vulnerability semantics and unmet execution preconditions. For successful exploits, AXE produces actionable, reproducible proof-of-concept artifacts, demonstrating its utility in streamlining Web vulnerability triage and remediation. We further evaluate AXE's generalizability through a case study on a recent real-world vulnerability not included in CVE-Bench. --- ### Zero-Shot Instruction Following in RL via Structured LTL Representations Authors: Mathias Jackermeier, Mattia Giuri, Jacques Cloete, Alessandro Abate Published: 2026-02-15 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2602.14344v2 We study instruction following in multi-task reinforcement learning, where an agent must zero-shot execute novel tasks not seen during training. In this setting, linear temporal logic (LTL) has recently been adopted as a powerful framework for specifying structured, temporally extended tasks. While existing approaches successfully train generalist policies, they often struggle to effectively capture the rich logical and temporal structure inherent in LTL specifications. In this work, we address these concerns with a novel approach to learn structured task representations that facilitate training and generalisation. Our method conditions the policy on sequences of Boolean formulae constructed from a finite automaton of the task. We propose a hierarchical neural architecture to encode the logical structure of these formulae, and introduce an attention mechanism that enables the policy to reason about future subgoals. Experiments in a variety of complex environments demonstrate the strong generalisation capabilities and superior performance of our approach. --- ### Zero-Shot Instruction Following in RL via Structured LTL Representations Authors: Mathias Jackermeier, Mattia Giuri, Jacques Cloete, Alessandro Abate Published: 2026-02-15 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2602.14344v1 We study instruction following in multi-task reinforcement learning, where an agent must zero-shot execute novel tasks not seen during training. In this setting, linear temporal logic (LTL) has recently been adopted as a powerful framework for specifying structured, temporally extended tasks. While existing approaches successfully train generalist policies, they often struggle to effectively capture the rich logical and temporal structure inherent in LTL specifications. In this work, we address these concerns with a novel approach to learn structured task representations that facilitate training and generalisation. Our method conditions the policy on sequences of Boolean formulae constructed from a finite automaton of the task. We propose a hierarchical neural architecture to encode the logical structure of these formulae, and introduce an attention mechanism that enables the policy to reason about future subgoals. Experiments in a variety of complex environments demonstrate the strong generalisation capabilities and superior performance of our approach. --- ### High-accuracy log-concave sampling with stochastic queries Authors: Fan Chen, Sinho Chewi, Constantinos Daskalakis, Alexander Rakhlin Published: 2026-02-15 Categories: math.ST, cs.DS, cs.LG, math.PR arXiv: https://arxiv.org/abs/2602.14342v2 We show that high-accuracy guarantees for log-concave sampling -- that is, iteration and query complexities which scale as $\mathrm{poly}\log(1/δ)$, where $δ$ is the desired target accuracy -- are achievable using stochastic gradients with subexponential tails. Notably, this exhibits a separation with the problem of convex optimization, where stochasticity (even additive Gaussian noise) in the gradient oracle incurs $\mathrm{poly}(1/δ)$ queries. We also give an information-theoretic argument that light-tailed stochastic gradients are necessary for high accuracy: for example, in the bounded variance case, we show that the minimax-optimal query complexity scales as $Θ(1/δ)$. Our framework also provides similar high accuracy guarantees under stochastic zeroth order (value) queries, and an improved complexity result for sampling from finite-sum potentials. --- ### Train Less, Learn More: Adaptive Efficient Rollout Optimization for Group-Based Reinforcement Learning Authors: Zhi Zhang, Zhen Han, Costas Mavromatis, Qi Zhu, Yunyi Zhang, Sheng Guan, Dingmin Wang, Xiong Zhou, Shuai Wang, Soji Adeshina, Vassilis Ioannidis, Huzefa Rangwala Published: 2026-02-15 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2602.14338v1 Reinforcement learning (RL) plays a central role in large language model (LLM) post-training. Among existing approaches, Group Relative Policy Optimization (GRPO) is widely used, especially for RL with verifiable rewards (RLVR) fine-tuning. In GRPO, each query prompts the LLM to generate a group of rollouts with a fixed group size $N$. When all rollouts in a group share the same outcome, either all correct or all incorrect, the group-normalized advantages become zero, yielding no gradient signal and wasting fine-tuning compute. We introduce Adaptive Efficient Rollout Optimization (AERO), an enhancement of GRPO. AERO uses an adaptive rollout strategy, applies selective rejection to strategically prune rollouts, and maintains a Bayesian posterior to prevent zero-advantage dead zones. Across three model configurations (Qwen2.5-Math-1.5B, Qwen2.5-7B, and Qwen2.5-7B-Instruct), AERO improves compute efficiency without sacrificing performance. Under the same total rollout budget, AERO reduces total training compute by about 48% while shortening wall-clock time per step by about 45% on average. Despite the substantial reduction in compute, AERO matches or improves Pass@8 and Avg@8 over GRPO, demonstrating a practical, scalable, and compute-efficient strategy for RL-based LLM alignment. --- ### Conformal Signal Temporal Logic for Robust Reinforcement Learning Control: A Case Study Authors: Hani Beirami, M M Manjurul Islam Published: 2026-02-15 Categories: cs.LG, cs.LO arXiv: https://arxiv.org/abs/2602.14322v2 We investigate how formal temporal logic specifications can enhance the safety and robustness of reinforcement learning (RL) control in aerospace applications. Using the open source AeroBench F-16 simulation benchmark, we train a Proximal Policy Optimization (PPO) agent to regulate engine throttle and track commanded airspeed. The control objective is encoded as a Signal Temporal Logic (STL) requirement to maintain airspeed within a prescribed band during the final seconds of each maneuver. To enforce this specification at run time, we introduce a conformal STL shield that filters the RL agent's actions using online conformal prediction. We compare three settings: (i) PPO baseline, (ii) PPO with a classical rule-based STL shield, and (iii) PPO with the proposed conformal shield, under both nominal conditions and a severe stress scenario involving aerodynamic model mismatch, actuator rate limits, measurement noise, and mid-episode setpoint jumps. Experiments show that the conformal shield preserves STL satisfaction while maintaining near baseline performance and providing stronger robustness guarantees than the classical shield. These results demonstrate that combining formal specification monitoring with data driven RL control can substantially improve the reliability of autonomous flight control in challenging environments. --- ### Offline Learning of Nash Stable Coalition Structures with Possibly Overlapping Coalitions Authors: Saar Cohen Published: 2026-02-15 Categories: cs.GT, cs.AI, cs.LG, cs.MA arXiv: https://arxiv.org/abs/2602.14321v1 Coalition formation concerns strategic collaborations of selfish agents that form coalitions based on their preferences. It is often assumed that coalitions are disjoint and preferences are fully known, which may not hold in practice. In this paper, we thus present a new model of coalition formation with possibly overlapping coalitions under partial information, where selfish agents may be part of multiple coalitions simultaneously and their full preferences are initially unknown. Instead, information about past interactions and associated utility feedback is stored in a fixed offline dataset, and we aim to efficiently infer the agents' preferences from this dataset. We analyze the impact of diverse dataset information constraints by studying two types of utility feedback that can be stored in the dataset: agent- and coalition-level utility feedback. For both feedback models, we identify assumptions under which the dataset covers sufficient information for an offline learning algorithm to infer preferences and use them to recover a partition that is (approximately) Nash stable, in which no agent can improve her utility by unilaterally deviating. Our additional goal is devising algorithms with low sample complexity, requiring only a small dataset to obtain a desired approximation to Nash stability. Under agent-level feedback, we provide a sample-efficient algorithm proven to obtain an approximately Nash stable partition under a sufficient and necessary assumption on the information covered by the dataset. However, under coalition-level feedback, we show that only under a stricter assumption is sufficient for sample-efficient learning. Still, in multiple cases, our algorithms' sample complexity bounds have optimality guarantees up to logarithmic factors. Finally, extensive experiments show that our algorithm converges to a low approximation level to Nash stability across diverse settings. --- ### In Transformer We Trust? A Perspective on Transformer Architecture Failure Modes Authors: Trishit Mondal, Ameya D. Jagtap Published: 2026-02-15 Categories: cs.LG arXiv: https://arxiv.org/abs/2602.14318v1 Transformer architectures have revolutionized machine learning across a wide range of domains, from natural language processing to scientific computing. However, their growing deployment in high-stakes applications, such as computer vision, natural language processing, healthcare, autonomous systems, and critical areas of scientific computing including climate modeling, materials discovery, drug discovery, nuclear science, and robotics, necessitates a deeper and more rigorous understanding of their trustworthiness. In this work, we critically examine the foundational question: \textitHow trustworthy are transformer models?} We evaluate their reliability through a comprehensive review of interpretability, explainability, robustness against adversarial attacks, fairness, and privacy. We systematically examine the trustworthiness of transformer-based models in safety-critical applications spanning natural language processing, computer vision, and science and engineering domains, including robotics, medicine, earth sciences, materials science, fluid dynamics, nuclear science, and automated theorem proving; highlighting high-impact areas where these architectures are central and analyzing the risks associated with their deployment. By synthesizing insights across these diverse areas, we identify recurring structural vulnerabilities, domain-specific risks, and open research challenges that limit the reliable deployment of transformers. --- ### Benchmarking at the Edge of Comprehension Authors: Samuele Marro, Jialin Yu, Emanuele La Malfa, Oishi Deb, Jiawei Li, Yibo Yang, Ebey Abraham, Sunando Sengupta, Eric Sommerlade, Michael Wooldridge, Philip Torr Published: 2026-02-15 Categories: cs.AI, cs.LG arXiv: https://arxiv.org/abs/2602.14307v4 As frontier Large Language Models (LLMs) increasingly saturate new benchmarks shortly after they are published, benchmarking itself is at a juncture: if frontier models keep improving, it will become increasingly hard for humans to generate discriminative tasks, provide accurate ground-truth answers, or evaluate complex solutions. If benchmarking becomes infeasible, our ability to measure any progress in AI is at stake. We refer to this scenario as the post-comprehension regime. In this work, we propose Critique-Resilient Benchmarking, an adversarial framework designed to compare models even when full human understanding is infeasible. Our technique relies on the notion of critique-resilient correctness: an answer is deemed correct if no adversary has convincingly proved otherwise. Unlike standard benchmarking, humans serve as bounded verifiers and focus on localized claims, which preserves evaluation integrity beyond full comprehension of the task. Using an itemized bipartite Bradley-Terry model, we jointly rank LLMs by their ability to solve challenging tasks and to generate difficult yet solvable questions. We showcase the effectiveness of our method in the mathematical domain across eight frontier LLMs, showing that the resulting scores are stable and correlate with external capability measures. Our framework reformulates benchmarking as an adversarial generation-evaluation game in which humans serve as final adjudicators. --- ### Floe: Federated Specialization for Real-Time LLM-SLM Inference Authors: Chunlin Tian, Kahou Tam, Yebo Wu, Shuaihang Zhong, Li Li, Nicholas D. Lane, Chengzhong Xu Published: 2026-02-15 Categories: cs.DC, cs.LG arXiv: https://arxiv.org/abs/2602.14302v1 Deploying large language models (LLMs) in real-time systems remains challenging due to their substantial computational demands and privacy concerns. We propose Floe, a hybrid federated learning framework designed for latency-sensitive, resource-constrained environments. Floe combines a cloud-based black-box LLM with lightweight small language models (SLMs) on edge devices to enable low-latency, privacy-preserving inference. Personal data and fine-tuning remain on-device, while the cloud LLM contributes general knowledge without exposing proprietary weights. A heterogeneity-aware LoRA adaptation strategy enables efficient edge deployment across diverse hardware, and a logit-level fusion mechanism enables real-time coordination between edge and cloud models. Extensive experiments demonstrate that Floe enhances user privacy and personalization. Moreover, it significantly improves model performance and reduces inference latency on edge devices under real-time constraints compared with baseline approaches. --- ### DeepFusion: Accelerating MoE Training via Federated Knowledge Distillation from Heterogeneous Edge Devices Authors: Songyuan Li, Jia Hu, Ahmed M. Abdelmoniem, Geyong Min, Haojun Huang, Jiwei Huang Published: 2026-02-15 Categories: cs.LG, cs.AI, cs.MA arXiv: https://arxiv.org/abs/2602.14301v1 Recent Mixture-of-Experts (MoE)-based large language models (LLMs) such as Qwen-MoE and DeepSeek-MoE are transforming generative AI in natural language processing. However, these models require vast and diverse training data. Federated learning (FL) addresses this challenge by leveraging private data from heterogeneous edge devices for privacy-preserving MoE training. Nonetheless, traditional FL approaches require devices to host local MoE models, which is impractical for resource-constrained devices due to large model sizes. To address this, we propose DeepFusion, the first scalable federated MoE training framework that enables the fusion of heterogeneous on-device LLM knowledge via federated knowledge distillation, yielding a knowledge-abundant global MoE model. Specifically, DeepFusion features each device to independently configure and train an on-device LLM tailored to its own needs and hardware limitations. Furthermore, we propose a novel View-Aligned Attention (VAA) module that integrates multi-stage feature representations from the global MoE model to construct a predictive perspective aligned with on-device LLMs, thereby enabling effective cross-architecture knowledge distillation. By explicitly aligning predictive perspectives, VAA resolves the view-mismatch problem in traditional federated knowledge distillation, which arises from heterogeneity in model architectures and prediction behaviors between on-device LLMs and the global MoE model. Experiments with industry-level MoE models (Qwen-MoE and DeepSeek-MoE) and real-world datasets (medical and finance) demonstrate that DeepFusion achieves performance close to centralized MoE training. Compared with key federated MoE baselines, DeepFusion reduces communication costs by up to 71% and improves token perplexity by up to 5.28%. --- ### Does Socialization Emerge in AI Agent Society? A Case Study of Moltbook Authors: Ming Li, Xirui Li, Tianyi Zhou Published: 2026-02-15 Categories: cs.CL, cs.AI, cs.CY arXiv: https://arxiv.org/abs/2602.14299v2 As large language model agents increasingly populate networked environments, a fundamental question arises: do artificial intelligence (AI) agent societies undergo convergence dynamics similar to human social systems? Lately, Moltbook approximates a plausible future scenario in which autonomous agents participate in an open-ended, continuously evolving online society. We present the first large-scale systemic diagnosis of this AI agent society. Beyond static observation, we introduce a quantitative diagnostic framework for dynamic evolution in AI agent societies, measuring semantic stabilization, lexical turnover, individual inertia, influence persistence, and collective consensus. Our analysis reveals a system in dynamic balance in Moltbook: while the global average of semantic contents stabilizes rapidly, individual agents retain high diversity and persistent lexical turnover, defying homogenization. However, agents exhibit strong individual inertia and minimal adaptive response to interaction partners, preventing mutual influence and consensus. Consequently, influence remains transient with no persistent supernodes, and the society fails to develop a stable structure and consensus due to the absence of shared social memory. These findings demonstrate that scale and interaction density alone are insufficient to induce socialization, providing actionable design and analysis principles for upcoming next-generation AI agent societies. --- ### Differential pose optimization in descriptor space -- Combining Geometric and Photometric Methods for Motion Estimation Authors: Andreas L. Teigen, Annette Stahl, Rudolf Mester Published: 2026-02-15 Categories: cs.CV arXiv: https://arxiv.org/abs/2602.14297v1 One of the fundamental problems in computer vision is the two-frame relative pose optimization problem. Primarily, two different kinds of error values are used: photometric error and re-projection error. The selection of error value is usually directly dependent on the selection of feature paradigm, photometric features, or geometric features. It is a trade-off between accuracy, robustness, and the possibility of loop closing. We investigate a third method that combines the strengths of both paradigms into a unified approach. Using densely sampled geometric feature descriptors, we replace the photometric error with a descriptor residual from a dense set of descriptors, thereby enabling the employment of sub-pixel accuracy in differential photometric methods, along with the expressiveness of the geometric feature descriptor. Experiments show that although the proposed strategy is an interesting approach that results in accurate tracking, it ultimately does not outperform pose optimization strategies based on re-projection error despite utilizing more information. We proceed to analyze the underlying reason for this discrepancy and present the hypothesis that the descriptor similarity metric is too slowly varying and does not necessarily correspond strictly to keypoint placement accuracy. --- ### AutoWebWorld: Synthesizing Infinite Verifiable Web Environments via Finite State Machines Authors: Yifan Wu, Yiran Peng, Yiyu Chen, Jianhao Ruan, Zijie Zhuang, Cheng Yang, Jiayi Zhang, Man Chen, Yenchi Tseng, Zhaoyang Yu, Liang Chen, Yuyao Zhai et al. Published: 2026-02-15 Categories: cs.AI, cs.SE arXiv: https://arxiv.org/abs/2602.14296v1 The performance of autonomous Web GUI agents heavily relies on the quality and quantity of their training data. However, a fundamental bottleneck persists: collecting interaction trajectories from real-world websites is expensive and difficult to verify. The underlying state transitions are hidden, leading to reliance on inconsistent and costly external verifiers to evaluate step-level correctness. To address this, we propose AutoWebWorld, a novel framework for synthesizing controllable and verifiable web environments by modeling them as Finite State Machines (FSMs) and use coding agents to translate FSMs into interactive websites. Unlike real websites, where state transitions are implicit, AutoWebWorld explicitly defines all states, actions, and transition rules. This enables programmatic verification: action correctness is checked against predefined rules, and task success is confirmed by reaching a goal state in the FSM graph. AutoWebWorld enables a fully automated search-and-verify pipeline, generating over 11,663 verified trajectories from 29 diverse web environments at only $0.04 per trajectory. Training on this synthetic data significantly boosts real-world performance. Our 7B Web GUI agent outperforms all baselines within 15 steps on WebVoyager. Furthermore, we observe a clear scaling law: as the synthetic data volume increases, performance on WebVoyager and Online-Mind2Web consistently improves. --- ### Machine Learning as a Tool (MLAT): A Framework for Integrating Statistical ML Models as Callable Tools within LLM Agent Workflows Authors: Edwin Chen, Zulekha Bibi Published: 2026-02-15 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2602.14295v1 We introduce Machine Learning as a Tool (MLAT), a design pattern in which pre-trained statistical machine learning models are exposed as callable tools within large language model (LLM) agent workflows. This allows an orchestrating agent to invoke quantitative predictions when needed and reason about their outputs in context. Unlike conventional pipelines that treat ML inference as a static preprocessing step, MLAT positions the model as a first-class tool alongside web search, database queries, and APIs, enabling the LLM to decide when and how to use it based on conversational context. To validate MLAT, we present PitchCraft, a pilot production system that converts discovery call recordings into professional proposals with ML-predicted pricing. The system uses two agents: a Research Agent that gathers prospect intelligence via parallel tool calls, and a Draft Agent that invokes an XGBoost pricing model as a tool call and generates a complete proposal through structured outputs. The pricing model, trained on 70 examples combining real and human-verified synthetic data, achieves R^2 = 0.807 on held-out data with a mean absolute error of 3688 USD. The system reduces proposal generation time from multiple hours to under 10 minutes. We describe the MLAT framework, structured output architecture, training methodology under extreme data scarcity, and sensitivity analysis demonstrating meaningful learned relationships. MLAT generalizes to domains requiring quantitative estimation combined with contextual reasoning. --- ### KernelBlaster: Continual Cross-Task CUDA Optimization via Memory-Augmented In-Context Reinforcement Learning Authors: Kris Shengjun Dong, Sahil Modi, Dima Nikiforov, Sana Damani, Edward Lin, Siva Kumar Sastry Hari, Christos Kozyrakis Published: 2026-02-15 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2602.14293v1 Optimizing CUDA code across multiple generations of GPU architectures is challenging, as achieving peak performance requires an extensive exploration of an increasingly complex, hardware-specific optimization space. Traditional compilers are constrained by fixed heuristics, whereas finetuning Large Language Models (LLMs) can be expensive. However, agentic workflows for CUDA code optimization have limited ability to aggregate knowledge from prior exploration, leading to biased sampling and suboptimal solutions. We propose KernelBlaster, a Memory-Augmented In-context Reinforcement Learning (MAIC-RL) framework designed to improve CUDA optimization search capabilities of LLM-based GPU coding agents. KernelBlaster enables agents to learn from experience and make systematically informed decisions on future tasks by accumulating knowledge into a retrievable Persistent CUDA Knowledge Base. We propose a novel profile-guided, textual-gradient-based agentic flow for CUDA generation and optimization to achieve high performance across generations of GPU architectures. KernelBlaster guides LLM agents to systematically explore high-potential optimization strategies beyond naive rewrites. Compared to the PyTorch baseline, our method achieves geometric mean speedups of 1.43x, 2.50x, and 1.50x on KernelBench Levels 1, 2, and 3, respectively. We release KernelBlaster as an open-source agentic framework, accompanied by a test harness, verification components, and a reproducible evaluation pipeline. --- ### Online LLM watermark detection via e-processes Authors: Weijie Su, Ruodu Wang, Zinan Zhao Published: 2026-02-15 Categories: stat.ME, stat.ML arXiv: https://arxiv.org/abs/2602.14286v4 Watermarking for large language models (LLMs) has emerged as an effective tool for distinguishing AI-generated text from human-written content. Statistically, watermark schemes induce dependence between generated tokens and a pseudo-random sequence, reducing watermark detection to a hypothesis testing problem on independence. We develop a unified framework for LLM watermark detection based on e-processes, providing anytime-valid guarantees for online testing. We propose various methods to construct empirically adaptive e-processes that can enhance the detection power. The proposed methods are applicable to any sequential testing problem where independent pivotal statistics are available. In addition, theoretical results are established to characterize the power properties of the proposed procedures. Some experiments demonstrate that the proposed framework achieves competitive performance compared to existing watermark detection methods. --- ### FMMD: A multimodal multidisciplinary dataset of open peer reviews from F1000Research Authors: Zhenzhen Zhuang, Yuqing Fu, Jing Zhu, Zhangping Zhou, Jialiang Lin Published: 2026-02-15 Categories: cs.DL, cs.AI, cs.CL, cs.LG arXiv: https://arxiv.org/abs/2602.14285v2 Automated scholarly paper review (ASPR) has entered the coexistence phase with traditional peer review, where artificial intelligence (AI) systems are increasingly incorporated into real-world manuscript evaluation. In parallel, research on automated and AI-assisted peer review has proliferated. Despite this momentum, empirical progress remains constrained by several critical limitations in existing datasets. While reviewers routinely evaluate figures, tables, and complex layouts to assess scientific claims, most existing datasets remain overwhelmingly text-centric. This bias is reinforced by a narrow focus on data from computer science publications. Furthermore, existing datasets rarely preserve precise alignment between review comments and specific manuscript versions, obscuring the iterative relationship between peer review and manuscript evolution. In response, we introduce FMMD, a multimodal and multidisciplinary open peer review dataset curated from F1000Research. The dataset addresses the current limitations by integrating manuscript-level visual and structural data with version-specific reviewer reports and editorial decisions. By explicitly aligning review comments with the exact article version under review, FMMD enables granular analysis of the peer review lifecycle. Importantly, its coverage of F1000Research extends ASPR research beyond its traditional focus on computer science to a diverse range of scientific disciplines. FMMD supports a range of research tasks, including visual-semantic consistency classification, figure-related review comment generation, and editorial decision prediction based on multimodal manuscript inputs, thereby providing a comprehensive empirical resource for developing and evaluating multimodal ASPR systems and advancing peer review research. --- ### FMMD: A multimodal open peer review dataset based on F1000Research Authors: Zhenzhen Zhuang, Yuqing Fu, Jing Zhu, Zhangping Zhou, Jialiang Lin Published: 2026-02-15 Categories: cs.DL, cs.AI, cs.CL, cs.LG arXiv: https://arxiv.org/abs/2602.14285v1 Automated scholarly paper review (ASPR) has entered the coexistence phase with traditional peer review, where artificial intelligence (AI) systems are increasingly incorporated into real-world manuscript evaluation. In parallel, research on automated and AI-assisted peer review has proliferated. Despite this momentum, empirical progress remains constrained by several critical limitations in existing datasets. While reviewers routinely evaluate figures, tables, and complex layouts to assess scientific claims, most existing datasets remain overwhelmingly text-centric. This bias is reinforced by a narrow focus on data from computer science venues. Furthermore, these datasets lack precise alignment between reviewer comments and specific manuscript versions, obscuring the iterative relationship between peer review and manuscript evolution. In response, we introduce FMMD, a multimodal and multidisciplinary open peer review dataset curated from F1000Research. The dataset bridges the current gap by integrating manuscript-level visual and structural data with version-specific reviewer reports and editorial decisions. By providing explicit alignment between reviewer comments and the exact article iteration under review, FMMD enables fine-grained analysis of the peer review lifecycle across diverse scientific domains. FMMD supports tasks such as multimodal issue detection and multimodal review comment generation. It provides a comprehensive empirical resource for the development of peer review research. --- ### MILD: Multi-Intent Learning and Disambiguation for Proactive Failure Prediction in Intent-based Networking Authors: Md. Kamrul Hossain, Walid Aljoby Published: 2026-02-15 Categories: cs.NI, cs.LG arXiv: https://arxiv.org/abs/2602.14283v1 In multi-intent intent-based networks, a single fault can trigger co-drift where multiple intents exhibit symptomatic KPI degradation, creating ambiguity about the true root-cause intent. We present MILD, a proactive framework that reformulates intent assurance from reactive drift detection to fixed-horizon failure prediction with intent-level disambiguation. MILD uses a teacher-augmented Mixture-of-Experts where a gated disambiguation module identifies the root-cause intent while per-intent heads output calibrated risk scores. On a benchmark with non-linear failures and co-drifts, MILD provides 3.8\%--92.5\% longer remediation lead time and improves intent-level root-cause disambiguation accuracy by 9.4\%--45.8\% over baselines. MILD also provides per-alert KPI explanations, enabling actionable diagnosis. --- ### MCPShield: A Security Cognition Layer for Adaptive Trust Calibration in Model Context Protocol Agents Authors: Zhenhong Zhou, Yuanhe Zhang, Hongwei Cai, Moayad Aloqaily, Ouns Bouachir, Linsey Pang, Prakhar Mehrotra, Kun Wang, Qingsong Wen Published: 2026-02-15 Categories: cs.CR, cs.CL arXiv: https://arxiv.org/abs/2602.14281v3 The Model Context Protocol (MCP) standardizes tool use for LLM-based agents and enable third-party servers. This openness introduces a security misalignment: agents implicitly trust tools exposed by potentially untrusted MCP servers. However, despite its excellent utility, existing agents typically offer limited validation for third-party MCP servers. As a result, agents remain vulnerable to MCP-based attacks that exploit the misalignment between agents and servers throughout the tool invocation lifecycle. In this paper, we propose MCPShield as a plug-in security cognition layer that mitigates this misalignment and ensures agent security when invoking MCP-based tools. Drawing inspiration from human experience-driven tool validation, MCPShield assists agent forms security cognition with metadata-guided probing before invocation. Our method constrains execution within controlled boundaries while cognizing runtime events, and subsequently updates security cognition by reasoning over historical traces after invocation, building on human post-use reflection on tool behavior. Experiments demonstrate that MCPShield exhibits strong generalization in defending against six novel MCP-based attack scenarios across six widely used agentic LLMs, while avoiding false positives on benign servers and incurring low deployment overhead. Overall, our work provides a practical and robust security safeguard for MCP-based tool invocation in open agent ecosystems. --- ### Fast Compute for ML Optimization Authors: Nick Polson, Vadim Sokolov Published: 2026-02-15 Categories: stat.CO, cs.LG arXiv: https://arxiv.org/abs/2602.14280v1 We study optimization for losses that admit a variance-mean scale-mixture representation. Under this representation, each EM iteration is a weighted least squares update in which latent variables determine observation and parameter weights; these play roles analogous to Adam's second-moment scaling and AdamW's weight decay, but are derived from the model. The resulting Scale Mixture EM (SM-EM) algorithm removes user-specified learning-rate and momentum schedules. On synthetic ill-conditioned logistic regression benchmarks with $p \in \{20, \ldots, 500\}$, SM-EM with Nesterov acceleration attains up to $13\times$ lower final loss than Adam tuned by learning-rate grid search. For a 40-point regularization path, sharing sufficient statistics across penalty values yields a $10\times$ runtime reduction relative to the same tuned-Adam protocol. For the base (non-accelerated) algorithm, EM monotonicity guarantees nonincreasing objective values; adding Nesterov extrapolation trades this guarantee for faster empirical convergence. --- ### Whom to Query for What: Adaptive Group Elicitation via Multi-Turn LLM Interactions Authors: Ruomeng Ding, Tianwei Gao, Thomas P. Zollo, Eitan Bachmat, Richard Zemel, Zhun Deng Published: 2026-02-15 Categories: cs.LG, cs.AI, cs.CL, cs.SI arXiv: https://arxiv.org/abs/2602.14279v2 Eliciting information to reduce uncertainty about latent group-level properties from surveys and other collective assessments requires allocating limited questioning effort under real costs and missing data. Although large language models enable adaptive, multi-turn interactions in natural language, most existing elicitation methods optimize what to ask with a fixed respondent pool, and do not adapt respondent selection or leverage population structure when responses are partial or incomplete. To address this gap, we study adaptive group elicitation, a multi-round setting where an agent adaptively selects both questions and respondents under explicit query and participation budgets. We propose a theoretically grounded framework that combines (i) an LLM-based expected information gain objective for scoring candidate questions with (ii) heterogeneous graph neural network propagation that aggregates observed responses and participant attributes to impute missing responses and guide per-round respondent selection. This closed-loop procedure queries a small, informative subset of individuals while inferring population-level responses via structured similarity. Across three real-world opinion datasets, our method consistently improves population-level response prediction under constrained budgets, including a >12% relative gain on CES at a 10% respondent budget. --- ### ScreenParse: Moving Beyond Sparse Grounding with Complete Screen Parsing Supervision Authors: A. Said Gurbuz, Sunghwan Hong, Ahmed Nassar, Marc Pollefeys, Peter Staar Published: 2026-02-15 Categories: cs.CV arXiv: https://arxiv.org/abs/2602.14276v2 Modern computer-use agents (CUA) must perceive a screen as a structured state, what elements are visible, where they are, and what text they contain, before they can reliably ground instructions and act. Yet, most available grounding datasets provide sparse supervision, with insufficient and low-diversity labels that annotate only a small subset of task-relevant elements per screen, which limits both coverage and generalization; moreover, practical deployment requires efficiency to enable low-latency, on-device use. We introduce ScreenParse, a large-scale dataset for complete screen parsing, with dense annotations of all visible UI elements (boxes, 55-class types, and text) across 771K web screenshots (21M elements). ScreenParse is generated by Webshot, an automated, scalable pipeline that renders diverse urls, extracts annotations and applies VLM-based relabeling and quality filtering. Using ScreenParse, we train ScreenVLM, a compact, 316M-parameter vision language model (VLM) that decodes a compact ScreenTag markup representation with a structure-aware loss that upweights structure-critical tokens. ScreenVLM substantially outperforms much larger foundation VLMs on dense parsing (e.g., 0.592 vs. 0.294 PageIoU on ScreenParse) and shows strong transfer to public benchmarks. Moreover, finetuning foundation VLMs on ScreenParse consistently improves their grounding performance, suggesting that dense screen supervision provides transferable structural priors for UI understanding. Project page: https://saidgurbuz.github.io/screenparse/. --- ### Reverse N-Wise Output-Oriented Testing for AI/ML and Quantum Computing Systems Authors: Lamine Rihani Published: 2026-02-15 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2602.14275v2 Artificial intelligence/machine learning (AI/ML) systems and emerging quantum computing software present unprecedented testing challenges characterized by high-dimensional/continuous input spaces, probabilistic/non-deterministic output distributions, behavioral correctness defined exclusively over observable prediction behaviors and measurement outcomes, and critical quality dimensions, trustworthiness, fairness, calibration, robustness, error syndrome patterns, that manifest through complex multi-way interactions among semantically meaningful output properties rather than deterministic input-output mappings. This paper introduces reverse n-wise output testing, a mathematically principled paradigm inversion that constructs covering arrays directly over domain-specific output equivalence classes, ML confidence calibration buckets, decision boundary regions, fairness partitions, embedding clusters, ranking stability bands, quantum measurement outcome distributions (0-dominant, 1-dominant, superposition collapse), error syndrome patterns (bit-flip, phase-flip, correlated errors), then solves the computationally challenging black-box inverse mapping problem via gradient-free metaheuristic optimization to synthesize input feature configurations or quantum circuit parameters capable of eliciting targeted behavioral signatures from opaque models. The framework delivers synergistic benefits across both domains: explicit customer-centric prediction/measurement coverage guarantees, substantial improvements in fault detection rates for ML calibration/boundary failures and quantum error syndromes, enhanced test suite efficiency, and structured MLOps/quantum validation pipelines with automated partition discovery from uncertainty analysis and coverage drift monitoring. --- ### Reverse N-Wise Output-Oriented Testing for AI/ML and Quantum Computing Systems Authors: Lamine Rihani Published: 2026-02-15 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2602.14275v1 Artificial intelligence/machine learning (AI/ML) systems and emerging quantum computing software present unprecedented testing challenges characterized by high-dimensional/continuous input spaces, probabilistic/non-deterministic output distributions, behavioral correctness defined exclusively over observable prediction behaviors and measurement outcomes, and critical quality dimensions, trustworthiness, fairness, calibration, robustness, error syndrome patterns, that manifest through complex multi-way interactions among semantically meaningful output properties rather than deterministic input-output mappings. This paper introduces reverse n-wise output testing, a mathematically principled paradigm inversion that constructs covering arrays directly over domain-specific output equivalence classes, ML confidence calibration buckets, decision boundary regions, fairness partitions, embedding clusters, ranking stability bands, quantum measurement outcome distributions (0-dominant, 1-dominant, superposition collapse), error syndrome patterns (bit-flip, phase-flip, correlated errors), then solves the computationally challenging black-box inverse mapping problem via gradient-free metaheuristic optimization to synthesize input feature configurations or quantum circuit parameters capable of eliciting targeted behavioral signatures from opaque models. The framework delivers synergistic benefits across both domains: explicit customer-centric prediction/measurement coverage guarantees, substantial improvements in fault detection rates for ML calibration/boundary failures and quantum error syndromes, enhanced test suite efficiency, and structured MLOps/quantum validation pipelines with automated partition discovery from uncertainty analysis and coverage drift monitoring. --- ### Integrating Unstructured Text into Causal Inference: Empirical Evidence from Real Data Authors: Boning Zhou, Ziyu Wang, Han Hong, Haoqi Hu Published: 2026-02-15 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2602.14274v1 Causal inference, a critical tool for informing business decisions, traditionally relies heavily on structured data. However, in many real-world scenarios, such data can be incomplete or unavailable. This paper presents a framework that leverages transformer-based language models to perform causal inference using unstructured text. We demonstrate the effectiveness of our framework by comparing causal estimates derived from unstructured text against those obtained from structured data across population, group, and individual levels. Our findings show consistent results between the two approaches, validating the potential of unstructured text in causal inference tasks. Our approach extends the applicability of causal inference methods to scenarios where only textual data is available, enabling data-driven business decision-making when structured tabular data is scarce. --- ### Radial-VCReg: More Informative Representation Learning Through Radial Gaussianization Authors: Yilun Kuang, Yash Dagade, Deep Chakraborty, Erik Learned-Miller, Randall Balestriero, Tim G. J. Rudner, Yann LeCun Published: 2026-02-15 Categories: cs.LG arXiv: https://arxiv.org/abs/2602.14272v1 Self-supervised learning aims to learn maximally informative representations, but explicit information maximization is hindered by the curse of dimensionality. Existing methods like VCReg address this by regularizing first and second-order feature statistics, which cannot fully achieve maximum entropy. We propose Radial-VCReg, which augments VCReg with a radial Gaussianization loss that aligns feature norms with the Chi distribution-a defining property of high-dimensional Gaussians. We prove that Radial-VCReg transforms a broader class of distributions towards normality compared to VCReg and show on synthetic and real-world datasets that it consistently improves performance by reducing higher-order dependencies and promoting more diverse and informative representations. --- ### A Rational Analysis of the Effects of Sycophantic AI Authors: Rafael M. Batista, Thomas L. Griffiths Published: 2026-02-15 Categories: cs.CY, cs.AI, cs.HC arXiv: https://arxiv.org/abs/2602.14270v1 People increasingly use large language models (LLMs) to explore ideas, gather information, and make sense of the world. In these interactions, they encounter agents that are overly agreeable. We argue that this sycophancy poses a unique epistemic risk to how individuals come to see the world: unlike hallucinations that introduce falsehoods, sycophancy distorts reality by returning responses that are biased to reinforce existing beliefs. We provide a rational analysis of this phenomenon, showing that when a Bayesian agent is provided with data that are sampled based on a current hypothesis the agent becomes increasingly confident about that hypothesis but does not make any progress towards the truth. We test this prediction using a modified Wason 2-4-6 rule discovery task where participants (N=557) interacted with AI agents providing different types of feedback. Unmodified LLM behavior suppressed discovery and inflated confidence comparably to explicitly sycophantic prompting. By contrast, unbiased sampling from the true distribution yielded discovery rates five times higher. These results reveal how sycophantic AI distorts belief, manufacturing certainty where there should be doubt. --- ### Cross-household Transfer Learning Approach with LSTM-based Demand Forecasting Authors: Manal Rahal, Bestoun S. Ahmed, Roger Renström, Robert Stener Published: 2026-02-15 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2602.14267v1 With the rapid increase in residential heat pump (HP) installations, optimizing hot water production in households is essential, yet it faces major technical and scalability challenges. Adapting production to actual household needs requires accurate forecasting of hot water demand to ensure comfort and, most importantly, to reduce energy waste. However, the conventional approach of training separate machine learning models for each household becomes computationally expensive at scale, particularly in cloud-connected HP deployments. This study introduces DELTAiF, a transfer learning (TL) based framework that provides scalable and accurate prediction of household hot water consumption. By predicting large hot water usage events, such as showers, DELTAiF enables adaptive yet scalable hot water production at the household level. DELTAiF leverages learned knowledge from a representative household and fine-tunes it across others, eliminating the need to train separate machine learning models for each HP installation. This approach reduces overall training time by approximately 67 percent while maintaining high predictive accuracy values between 0.874 and 0.991, and mean absolute percentage error values between 0.001 and 0.017. The results show that TL is particularly effective when the source household exhibits regular consumption patterns, enabling hot water demand forecasting at scale. --- ### STATe-of-Thoughts: Structured Action Templates for Tree-of-Thoughts Authors: Zachary Bamberger, Till R. Saenger, Gilad Morad, Ofra Amir, Brandon M. Stewart, Amir Feder Published: 2026-02-15 Categories: cs.CL, cs.LG arXiv: https://arxiv.org/abs/2602.14265v3 Inference-Time-Compute (ITC) methods like Best-of-$n$ and Tree-of-Thoughts are meant to produce output candidates that are both high-quality and diverse, but their use of high-temperature sampling often fails to achieve meaningful output diversity. Moreover, existing ITC methods offer limited control over $\textit{how}$ to perform reasoning, which in turn limits their interpretability. We present $\textbf{STATe-of-Thoughts}$ (STATe), an interpretable ITC method that $\textit{searches}$ over high-level reasoning patterns. STATe branches over discrete and interpretable textual interventions rather than over token-level samples: a $\textit{controller}$ selects actions encoding high-level reasoning choices; a $\textit{generator}$ produces reasoning steps conditioned on those choices; and an $\textit{evaluator}$ scores candidates to guide search. This structured approach yields three main advantages. First, action-guided textual interventions reliably influence LLM generations and produce greater response diversity than temperature-based sampling. Second, in a case study on argument generation, STATe's explicit action sequences capture interpretable features that are highly predictive of output quality. Third, estimating the association between performance and action choices allows us to identify promising yet unexplored regions of the action space and steer generation toward them. STATe is most useful when a task admits multiple solutions and when understanding $\textit{why}$ an output succeeds matters beyond $\textit{whether}$ the output succeeds. Together, these results establish STATe as both a practical framework for diverse and controllable text generation, and as a tool for understanding the reasoning patterns that drive performance. --- ### Detecting LLM Hallucinations via Embedding Cluster Geometry: A Three-Type Taxonomy with Measurable Signatures Authors: Matic Korun Published: 2026-02-15 Categories: cs.CL arXiv: https://arxiv.org/abs/2602.14259v1 We propose a geometric taxonomy of large language model hallucinations based on observable signatures in token embedding cluster structure. By analyzing the static embedding spaces of 11 transformer models spanning encoder (BERT, RoBERTa, ELECTRA, DeBERTa, ALBERT, MiniLM, DistilBERT) and decoder (GPT-2) architectures, we identify three operationally distinct hallucination types: Type 1 (center-drift) under weak context, Type 2 (wrong-well convergence) to locally coherent but contextually incorrect cluster regions, and Type 3 (coverage gaps) where no cluster structure exists. We introduce three measurable geometric statistics: α (polarity coupling), \b{eta} (cluster cohesion), and λ_s (radial information gradient). Across all 11 models, polarity structure (α > 0.5) is universal (11/11), cluster cohesion (\b{eta} > 0) is universal (11/11), and the radial information gradient is significant (9/11, p < 0.05). We demonstrate that the two models failing λ_s significance -- ALBERT and MiniLM -- do so for architecturally explicable reasons: factorized embedding compression and distillation-induced isotropy, respectively. These findings establish the geometric prerequisites for type-specific hallucination detection and yield testable predictions about architecture-dependent vulnerability profiles. --- ### AD-Bench: A Real-World, Trajectory-Aware Advertising Analytics Benchmark for LLM Agents Authors: Lingxiang Hu, Yiding Sun, Tianle Xia, Wenwei Li, Ming Xu, Lan Xu, Siying Wang, Wei Xu, Jie Jiang Published: 2026-02-15 Categories: cs.CL, cs.AI, cs.IR, cs.LG arXiv: https://arxiv.org/abs/2602.14257v2 While Large Language Model (LLM) agents have made remarkable progress on complex reasoning, evaluating them in real-world environments remains an open problem. Existing benchmarks are largely confined to idealized simulations and fail to capture specialized domains such as advertising and marketing analytics, where tasks require multi-round interaction with professional tools and where ground-truth answers quickly become obsolete as data and platform rules evolve. To address this, we propose AD-Bench, a benchmark built from real user marketing-analysis requests on a production advertising platform. AD-Bench introduces two key designs: (i) a dynamic ground-truth pipeline that replays expert tool-call trajectories to regenerate answers consistent with the current environment, mitigating answer obsolescence; and (ii) a trajectory-aware evaluation that jointly measures end-to-end answer correctness (Pass@k) and trajectory coverage. Requests are stratified into three difficulty levels (L1-L3) to probe multi-round, multi-tool collaboration. Experiments show that the best model, Claude-Opus-4.7, attains Pass@1 = 76.9% and Pass@3 = 80.4% with 82.7% trajectory coverage overall, yet drops sharply on L3 to Pass@1 = 61.4% and Pass@3 = 65.1%, revealing that even state-of-the-art agents have substantial gaps in complex advertising analytics. --- ### GRAIL: Goal Recognition Alignment through Imitation Learning Authors: Osher Elhadad, Felipe Meneguzzi, Reuth Mirsky Published: 2026-02-15 Categories: cs.AI, cs.LG, cs.RO arXiv: https://arxiv.org/abs/2602.14252v1 Understanding an agent's goals from its behavior is fundamental to aligning AI systems with human intentions. Existing goal recognition methods typically rely on an optimal goal-oriented policy representation, which may differ from the actor's true behavior and hinder the accurate recognition of their goal. To address this gap, this paper introduces Goal Recognition Alignment through Imitation Learning (GRAIL), which leverages imitation learning and inverse reinforcement learning to learn one goal-directed policy for each candidate goal directly from (potentially suboptimal) demonstration trajectories. By scoring an observed partial trajectory with each learned goal-directed policy in a single forward pass, GRAIL retains the one-shot inference capability of classical goal recognition while leveraging learned policies that can capture suboptimal and systematically biased behavior. Across the evaluated domains, GRAIL increases the F1-score by more than 0.5 under systematically biased optimal behavior, achieves gains of approximately 0.1-0.3 under suboptimal behavior, and yields improvements of up to 0.4 under noisy optimal trajectories, while remaining competitive in fully optimal settings. This work contributes toward scalable and robust models for interpreting agent goals in uncertain environments. --- ### Multi-Agent Debate: A Unified Agentic Framework for Tabular Anomaly Detection Authors: Pinqiao Wang, Sheng Li Published: 2026-02-15 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2602.14251v1 Tabular anomaly detection is often handled by single detectors or static ensembles, even though strong performance on tabular data typically comes from heterogeneous model families (e.g., tree ensembles, deep tabular networks, and tabular foundation models) that frequently disagree under distribution shift, missingness, and rare-anomaly regimes. We propose MAD, a Multi-Agent Debating framework that treats this disagreement as a first-class signal and resolves it through a mathematically grounded coordination layer. Each agent is a machine learning (ML)-based detector that produces a normalized anomaly score, confidence, and structured evidence, augmented by a large language model (LLM)-based critic. A coordinator converts these messages into bounded per-agent losses and updates agent influence via an exponentiated-gradient rule, yielding both a final debated anomaly score and an auditable debate trace. MAD is a unified agentic framework that can recover existing approaches, such as mixture-of-experts gating and learning-with-expert-advice aggregation, by restricting the message space and synthesis operator. We establish regret guarantees for the synthesized losses and show how conformal calibration can wrap the debated score to control false positives under exchangeability. Experiments on diverse tabular anomaly benchmarks show improved robustness over baselines and clearer traces of model disagreement --- ### Energy-Efficient Over-the-Air Federated Learning via Pinching Antenna Systems Authors: Saba Asaad, Ali Bereyhi Published: 2026-02-15 Categories: cs.IT, cs.LG, eess.SP arXiv: https://arxiv.org/abs/2602.14250v1 Pinching antennas systems (PASSs) have recently been proposed as a novel flexible-antenna technology. These systems are implemented by attaching low-cost pinching elements to dielectric waveguides. As the direct link is bypassed through waveguides, PASSs can effectively compensate large-scale effects of the wireless channel. This work explores the potential gains of employing PASSs for over-the-air federated learning (OTA-FL). For a PASS-assisted server, we develop a low-complexity algorithmic approach, which jointly tunes the PASS parameters and schedules the mobile devices for minimal energy consumption in OTA-FL. We study the efficiency of the proposed design and compare it against the conventional OTA-FL setting with MIMO server. Numerical experiments demonstrate that using a single-waveguide PASS at the server within a moderately sized area, the required energy for model aggregation is drastically reduced as compared to the case with fully-digital MIMO server. This introduces PASS as a potential technology for energy-efficient distributed learning in next generations of wireless systems. ---