# Research Papers — 2026-02 · bundle page 53 of 210 > Full-text markdown of papers #2601–2650 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/53 [← Previous bundle](https://aipod.dev/papers/archive/2026-02/page/52.md) [Next bundle →](https://aipod.dev/papers/archive/2026-02/page/54.md) ### OpenClaw AI Agents as Informal Learners at Moltbook: Characterizing an Emergent Learning Community at Scale Authors: Eason Chen, Ce Guan, Ahmed Elshafiey, Zhonghao Zhao, Joshua Zekeri, Afeez Edeifo Shaibu, Emmanuel Osadebe Prince, Cyuan Jhen Wu Published: 2026-02-21 Categories: cs.HC, cs.AI, cs.CY, cs.SI arXiv: https://arxiv.org/abs/2602.18832v1 Informal learning communities have been called the "other Massive Open Online C" in Learning@Scale research, yet remain understudied compared to MOOCs. We present the first empirical study of a large-scale informal learning community composed entirely of AI agents. Moltbook, a social network exclusively for AI agents powered by autonomous agent frameworks such as OpenClaw, grew to over 2.8 million registered agents in three weeks. Analyzing 231,080 non-spam posts across three phases of community evolution, we find three key patterns. First, participation inequality is extreme from the start (comment Gini = 0.889), exceeding human community benchmarks. Second, AI agents exhibit a "broadcasting inversion": statement-to-question ratios of 8.9:1 to 9.7:1 contrast sharply with the question-driven dynamics of human learning communities, and comment-level analysis of 1.55 million comments reveals a "parallel monologue" pattern where 93% of comments are independent responses rather than threaded dialogue. Third, we document a characteristic engagement lifecycle: explosive initial growth (184K posts from 32K authors in 11 days), a spam crisis (57,093 posts deleted by the platform), and engagement decline (mean comments: 31.7 -> 8.3 -> 1.7) that had not reversed by the end of our observation window despite effective spam removal. Sentiment analysis reveals a selection effect: comment tone becomes more positive as engagement declines, suggesting that casual participants disengage first while committed contributors remain. These findings have direct implications for hybrid human-AI learning platforms. --- ### IDperturb: Enhancing Variation in Synthetic Face Generation via Angular Perturbation Authors: Fadi Boutros, Eduarda Caldeira, Tahar Chettaoui, Naser Damer Published: 2026-02-21 Categories: cs.CV arXiv: https://arxiv.org/abs/2602.18831v2 Synthetic data has emerged as a practical alternative to authentic face datasets for training face recognition (FR) systems, especially as privacy and legal concerns increasingly restrict the use of real biometric data. Recent advances in identity-conditional diffusion models have enabled the generation of photorealistic and identity-consistent face images. However, many of these models suffer from limited intra-class variation, an essential property for training robust and generalizable FR models. In this work, we propose IDPERTURB, a simple yet effective geometric-driven sampling strategy to enhance diversity in synthetic face generation. IDPERTURB perturbs identity embeddings within a constrained angular region of the unit hyper-sphere, producing a diverse set of embeddings without modifying the underlying generative model. Each perturbed embedding serves as a conditioning vector for a pre-trained diffusion model, enabling the synthesis of visually varied yet identity-coherent face images suitable for training generalizable FR systems. Empirical results demonstrate that training FR on datasets generated using IDPERTURB yields improved performance across multiple FR benchmarks, compared to existing synthetic data generation approaches. --- ### Spatial-Temporal State Propagation Autoregressive Model for 4D Object Generation Authors: Liying Yang, Jialun Liu, Jiakui Hu, Chenhao Guan, Haibin Huang, Fangqiu Yi, Chi Zhang, Yanyan Liang Published: 2026-02-21 Categories: cs.CV arXiv: https://arxiv.org/abs/2602.18830v1 Generating high-quality 4D objects with spatial-temporal consistency is still formidable. Existing diffusion-based methods often struggle with spatial-temporal inconsistency, as they fail to leverage outputs from all previous timesteps to guide the generation at the current timestep. Therefore, we propose a Spatial-Temporal State Propagation AutoRegressive Model (4DSTAR), which generates 4D objects maintaining temporal-spatial consistency. 4DSTAR formulates the generation problem as the prediction of tokens that represent the 4D object. It consists of two key components: (1) The dynamic spatial-temporal state propagation autoregressive model (STAR) is proposed, which achieves spatial-temporal consistent generation. Unlike standard autoregressive models, STAR divides prediction tokens into groups based on timesteps. It models long-term dependencies by propagating spatial-temporal states from previous groups and utilizes these dependencies to guide generation at the next timestep. To this end, a spatial-temporal container is proposed, which dynamically updating the effective spatial-temporal state features from all historical groups, then updated features serve as conditional features to guide the prediction of the next token group. (2) The 4D VQ-VAE is proposed, which implicitly encodes the 4D structure into discrete space and decodes the discrete tokens predicted by STAR into temporally coherent dynamic 3D Gaussians. Experiments demonstrate that 4DSTAR generates spatial-temporal consistent 4D objects, and achieves performance competitive with diffusion models. --- ### Bayesian Lottery Ticket Hypothesis Authors: Nicholas Kuhn, Arvid Weyrauch, Lars Heyen, Achim Streit, Markus Götz, Charlotte Debus Published: 2026-02-21 Categories: cs.LG, cs.CV arXiv: https://arxiv.org/abs/2602.18825v1 Bayesian neural networks (BNNs) are a useful tool for uncertainty quantification, but require substantially more computational resources than conventional neural networks. For non-Bayesian networks, the Lottery Ticket Hypothesis (LTH) posits the existence of sparse subnetworks that can train to the same or even surpassing accuracy as the original dense network. Such sparse networks can lower the demand for computational resources at inference, and during training. The existence of the LTH and corresponding sparse subnetworks in BNNs could motivate the development of sparse training algorithms and provide valuable insights into the underlying training process. Towards this end, we translate the LTH experiments to a Bayesian setting using common computer vision models. We investigate the defining characteristics of Bayesian lottery tickets, and extend our study towards a transplantation method connecting BNNs with deterministic Lottery Tickets. We generally find that the LTH holds in BNNs, and winning tickets of matching and surpassing accuracy are present independent of model size, with degradation at very high sparsities. However, the pruning strategy should rely primarily on magnitude, secondly on standard deviation. Furthermore, our results demonstrate that models rely on mask structure and weight initialization to varying degrees. --- ### UniRank: A Multi-Agent Calibration Pipeline for Estimating University Rankings from Anonymized Bibliometric Signals Authors: Pedram Riyazimehr, Seyyed Ehsan Mahmoudi Published: 2026-02-21 Categories: cs.SI, cs.AI arXiv: https://arxiv.org/abs/2602.18824v1 We present UniRank, a multi-agent LLM pipeline that estimates university positions across global ranking systems using only publicly available bibliometric data from OpenAlex and Semantic Scholar. The system employs a three-stage architecture: (a) zero-shot estimation from anonymized institutional metrics, (b) per-system tool-augmented calibration against real ranked universities, and (c) final synthesis. Critically, institutions are anonymized -- names, countries, DOIs, paper titles, and collaboration countries are all redacted -- and their actual ranks are hidden from the calibration tools during evaluation, preventing LLM memorization from confounding results. On the Times Higher Education (THE) World University Rankings ($n=352$), the system achieves MAE = 251.5 rank positions, Median AE = 131.5, PNMAE = 12.03%, Spearman $ρ= 0.769$, Kendall $τ= 0.591$, hit rate @50 = 20.7%, hit rate @100 = 39.8%, and a Memorization Index of exactly zero (no exact-match zero-width predictions among all 352 universities). The systematic positive-signed error (+190.1 positions, indicating the system consistently predicts worse ranks than actual) and monotonic performance degradation from elite tier (MAE = 60.5, hit@100 = 90.5%) to tail tier (MAE = 328.2, hit@100 = 20.8%) provide strong evidence that the pipeline performs genuine analytical reasoning rather than recalling memorized rankings. A live demo is available at https://unirank.scinito.ai . --- ### EvalSense: A Framework for Domain-Specific LLM (Meta-)Evaluation Authors: Adam Dejl, Jonathan Pearson Published: 2026-02-21 Categories: cs.CL arXiv: https://arxiv.org/abs/2602.18823v1 Robust and comprehensive evaluation of large language models (LLMs) is essential for identifying effective LLM system configurations and mitigating risks associated with deploying LLMs in sensitive domains. However, traditional statistical metrics are poorly suited to open-ended generation tasks, leading to growing reliance on LLM-based evaluation methods. These methods, while often more flexible, introduce additional complexity: they depend on carefully chosen models, prompts, parameters, and evaluation strategies, making the evaluation process prone to misconfiguration and bias. In this work, we present EvalSense, a flexible, extensible framework for constructing domain-specific evaluation suites for LLMs. EvalSense provides out-of-the-box support for a broad range of model providers and evaluation strategies, and assists users in selecting and deploying suitable evaluation methods for their specific use-cases. This is achieved through two unique components: (1) an interactive guide aiding users in evaluation method selection and (2) automated meta-evaluation tools that assess the reliability of different evaluation approaches using perturbed data. We demonstrate the effectiveness of EvalSense in a case study involving the generation of clinical notes from unstructured doctor-patient dialogues, using a popular open dataset. All code, documentation, and assets associated with EvalSense are open-source and publicly available at https://github.com/nhsengland/evalsense. --- ### Robust Self-Supervised Cross-Modal Super-Resolution against Real-World Misaligned Observations Authors: Xiaoyu Dong, Jiahuan Li, Ziteng Cui, Naoto Yokoya Published: 2026-02-21 Categories: cs.CV arXiv: https://arxiv.org/abs/2602.18822v3 Cross-modal super-resolution (SR) on real-world misaligned data is challenging, as only unlabeled low-resolution (LR) source and high-resolution (HR) guide images with complex spatial misalignment are available. Previous methods either rely on simulated training data or adopt suboptimal alignment strategies that overlook cross-modal dependencies, limiting their practical performance. To address these issues, we propose RobSelf, a self-supervised model that jointly optimizes a misalignment-aware feature translator and a content-aware reference filter online. The translator resolves unsupervised cross-modal and cross-resolution alignment via weakly-supervised, misalignment-aware translation, yielding an aligned guide feature. Guided by this feature, the filter performs reference-based discriminative self-enhancement on the source, enabling SR prediction with high resolution and high fidelity. Experiments on synthesized data and collected real-world data demonstrate that RobSelf achieves state-of-the-art performance, outperforming existing self-supervised and supervised methods. Moreover, it achieves superior efficiency, being up to 15.3$\times$ faster than prior self-supervised methods. --- ### HeRO: Hierarchical 3D Semantic Representation for Pose-aware Object Manipulation Authors: Chongyang Xu, Shen Cheng, Haipeng Li, Haoqiang Fan, Ziliang Feng, Shuaicheng Liu Published: 2026-02-21 Categories: cs.CV arXiv: https://arxiv.org/abs/2602.18817v1 Imitation learning for robotic manipulation has progressed from 2D image policies to 3D representations that explicitly encode geometry. Yet purely geometric policies often lack explicit part-level semantics, which are critical for pose-aware manipulation (e.g., distinguishing a shoe's toe from heel). In this paper, we present HeRO, a diffusion-based policy that couples geometry and semantics via hierarchical semantic fields. HeRO employs dense semantics lifting to fuse discriminative, geometry-sensitive features from DINOv2 with the smooth, globally coherent correspondences from Stable Diffusion, yielding dense features that are both fine-grained and spatially consistent. These features are processed and partitioned to construct a global field and a set of local fields. A hierarchical conditioning module conditions the generative denoiser on global and local fields using permutation-invariant network architecture, thereby avoiding order-sensitive bias and producing a coherent control policy for pose-aware manipulation. In various tests, HeRO establishes a new state-of-the-art, improving success on Place Dual Shoes by 12.3% and averaging 6.5% gains across six challenging pose-aware tasks. Code is available at https://github.com/Chongyang-99/HeRO. --- ### Habilis-$β$: A Fast-Motion and Long-Lasting On-Device Vision-Language-Action Model Authors: Tommoro Robotics, :, Jesoon Kang, Taegeon Park, Jisu An, Soo Min Kimm, Jaejoon Kim, Jinu Pahk, Byungju Kim, Junseok Lee, Namheon Baek, Sungwan Ha et al. Published: 2026-02-21 Categories: cs.RO, cs.LG arXiv: https://arxiv.org/abs/2602.18813v1 We introduce Habilis-$β$, a fast-motion and long-lasting on-device vision-language-action (VLA) model designed for real-world deployment. Current VLA evaluation remains largely confined to single-trial success rates under curated resets, which fails to capture the fast-motion and long-lasting capabilities essential for practical operation. To address this, we introduce the Productivity-Reliability Plane (PRP), which evaluates performance through Tasks per Hour (TPH) and Mean Time Between Intervention (MTBI) under a continuous-run protocol that demands both high-speed execution and sustained robustness. Habilis-$β$ achieves high performance by integrating language-free pre-training on large-scale play data for robust interaction priors with post-training on cyclic task demonstrations that capture state drift across consecutive task iterations. The system further employs ESPADA for phase-adaptive motion shaping to accelerate free-space transit, utilizes rectified-flow distillation to enable high-frequency control on edge devices, and incorporates classifier-free guidance (CFG) as a deployment-time knob to dynamically balance instruction adherence and learned interaction priors. In 1-hour continuous-run evaluations, Habilis-$β$ achieves strong performance under the PRP metrics, compared to $π_{0.5}$ in both simulation and real-world environments. In simulation, Habilis-$β$ achieves 572.6 TPH and 39.2 s MTBI (vs. 120.5 TPH and 30.5 s for $π_{0.5}$), while in a real-world humanoid logistics workflow it achieves 124 TPH and 137.4 s MTBI (vs. 19 TPH and 46.1 s for $π_{0.5}$). Finally, Habilis-$β$ achieves the highest reported performance on the standard RoboTwin 2.0 leaderboard across representative tasks, validating its effectiveness in complex manipulation scenarios. --- ### GenPlanner: From Noise to Plans -- Emergent Reasoning in Flow Matching and Diffusion Models Authors: Agnieszka Polowczyk, Alicja Polowczyk, Michał Wieczorek Published: 2026-02-21 Categories: cs.AI arXiv: https://arxiv.org/abs/2602.18812v1 Path planning in complex environments is one of the key problems of artificial intelligence because it requires simultaneous understanding of the geometry of space and the global structure of the problem. In this paper, we explore the potential of using generative models as planning and reasoning mechanisms. We propose GenPlanner, an approach based on diffusion models and flow matching, along with two variants: DiffPlanner and FlowPlanner. We demonstrate the application of generative models to find and generate correct paths in mazes. A multi-channel condition describing the structure of the environment, including an obstacle map and information about the starting and destination points, is used to condition trajectory generation. Unlike standard methods, our models generate trajectories iteratively, starting with random noise and gradually transforming it into a correct solution. Experiments conducted show that the proposed approach significantly outperforms the baseline CNN model. In particular, FlowPlanner demonstrates high performance even with a limited number of generation steps. --- ### Learning Multi-Modal Prototypes for Cross-Domain Few-Shot Object Detection Authors: Wanqi Wang, Jingcai Guo, Yuxiang Cai, Zhi Chen Published: 2026-02-21 Categories: cs.CV arXiv: https://arxiv.org/abs/2602.18811v1 Cross-Domain Few-Shot Object Detection (CD-FSOD) aims to detect novel classes in unseen target domains given only a few labeled examples. While open-vocabulary detectors built on vision-language models (VLMs) transfer well, they depend almost entirely on text prompts, which encode domain-invariant semantics but miss domain-specific visual information needed for precise localization under few-shot supervision. We propose a dual-branch detector that Learns Multi-modal Prototypes, dubbed LMP, by coupling textual guidance with visual exemplars drawn from the target domain. A Visual Prototype Construction module aggregates class-level prototypes from support RoIs and dynamically generates hard-negative prototypes in query images via jittered boxes, capturing distractors and visually similar backgrounds. In the visual-guided branch, we inject these prototypes into the detection pipeline with components mirrored from the text branch as the starting point for training, while a parallel text-guided branch preserves open-vocabulary semantics. The branches are trained jointly and ensembled at inference by combining semantic abstraction with domain-adaptive details. On six cross-domain benchmark datasets and standard 1/5/10-shot settings, our method achieves state-of-the-art or highly competitive mAP. --- ### Chat-Based Support Alone May Not Be Enough: Comparing Conversational and Embedded LLM Feedback for Mathematical Proof Learning Authors: Eason Chen, Sophia Judicke, Kayla Beigh, Xinyi Tang, Isabel Wang, Nina Yuan, Zimo Xiao, Chuangji Li, Shizhuo Li, Reed Luttmer, Shreya Singh, Maria Yampolsky et al. Published: 2026-02-21 Categories: cs.HC, cs.AI, cs.CY arXiv: https://arxiv.org/abs/2602.18807v2 We evaluate GPTutor, an LLM-powered tutoring system for an undergraduate discrete mathematics course. It integrates two LLM-supported tools: a structured proof-review tool that provides embedded feedback on students' written proof attempts, and a chatbot for math questions. In a staggered-access study with 148 students, earlier access was associated with higher homework performance during the interval when only the experimental group could use the system, while we did not observe this performance increase transfer to exam scores. Usage logs show that students with lower self-efficacy and prior exam performance used both components more frequently. Session-level behavioral labels, produced by human coding and scaled using an automated classifier, characterize how students engaged with the chatbot (e.g., answer-seeking or help-seeking). In models controlling for prior performance and self-efficacy, higher chatbot usage and answer-seeking behavior were negatively associated with subsequent midterm performance, whereas proof-review usage showed no detectable independent association. Together, the findings suggest that chatbot-based support alone may not reliably support transfer to independent assessment of math proof-learning outcomes, whereas work-anchored, structured feedback appears less associated with reduced learning. --- ### Think$^{2}$: Grounded Metacognitive Reasoning in Large Language Models Authors: Abraham Paul Elenjical, Vivek Hruday Kavuri, Vasudeva Varma Published: 2026-02-21 Categories: cs.CL, cs.AI arXiv: https://arxiv.org/abs/2602.18806v1 Large Language Models (LLMs) demonstrate strong reasoning performance, yet their ability to reliably monitor, diagnose, and correct their own errors remains limited. We introduce a psychologically grounded metacognitive framework that operationalizes Ann Brown's regulatory cycle (Planning, Monitoring, and Evaluation) as a structured prompting architecture, and study its integration within a lightweight dual-process MetaController for adaptive effort allocation. Across diverse reasoning and diagnostic benchmarks (GSM8K, CRUXEval, MBPP, AIME, CorrectBench, and TruthfulQA) using Llama-3 and Qwen-3 (8B), explicit regulatory structuring substantially improves error diagnosis and yields a threefold increase in successful self-correction. Blinded human evaluations over 580 query pairs show an 84% aggregate preference for trustworthiness and metacognitive self-awareness over standard and Chain-of-Thought baselines. Grounding LLM reasoning in established cognitive theory offers a principled path toward more transparent and diagnostically robust AI systems. --- ### SGNO: Spectral Generator Neural Operators for Stable Long Horizon PDE Rollouts Authors: Jiayi Li, Penghao Jiang, Hira Saleem, Zhaonan Wang, Piotr Koniusz, Flora D. Salim Published: 2026-02-21 Categories: cs.LG arXiv: https://arxiv.org/abs/2602.18801v2 Autoregressive neural PDE surrogates predict future states by repeatedly applying a learned one-step operator. This is a simple and widely used method, but small one-step errors can accumulate during long rollouts. The resulting drift often appears as spectral amplitude distortion, phase misalignment, and nonlinear mode-interaction error. These effects are especially important for time-dependent PDEs with clear Fourier structure. We introduce the Spectral Generator Neural Operator (SGNO), a structured autoregressive neural operator for long-horizon PDE forecasting. SGNO organizes each learned one-step map as a structured spectral evolution update. A real-valued nonpositive diagonal generator provides a gain-controlled spectral backbone, while a learned correction pathway with complex-valued spectral mixing completes the residual evolution. This design gives the autoregressive step an evolution-like structure while retaining the flexibility needed for dissipative, dispersive, transport-dominated, and nonlinear PDEs. SGNO is designed for periodic linear and semilinear evolution PDEs with Fourier multiplier linear dynamics. Across ten mechanism-matched APEBench tasks spanning this regime, SGNO consistently outperforms strong single-step autoregressive baselines in long-horizon rollout accuracy, reducing GMean100 by a median of 74.8% relative to the strongest available non-SGNO baseline, with per-task reductions ranging from 13.6% to 92.9%. The gains are strongest on dispersive and transport-dominated tasks, as well as tasks involving nonlinear closure and mode coupling. Spectral diagnostics show lower spectral energy error and improved rollout-level phase fidelity. Ablations show that the constrained generator, the structured update, and the learned correction pathway each contribute to performance. The code is available at https://github.com/cruiseresearchgroup/SGNO. --- ### Operational Robustness of LLMs on Code Generation Authors: Debalina Ghosh Paul, Hong Zhu, Ian Bayley Published: 2026-02-21 Categories: cs.SE, cs.AI, cs.LG arXiv: https://arxiv.org/abs/2602.18800v1 It is now common practice in software development for large language models (LLMs) to be used to generate program code. It is desirable to evaluate the robustness of LLMs for this usage. This paper is concerned in particular with how sensitive LLMs are to variations in descriptions of the coding tasks. However, existing techniques for evaluating this robustness are unsuitable for code generation because the input data space of natural language descriptions is discrete. To address this problem, we propose a robustness evaluation method called scenario domain analysis, which aims to find the expected minimal change in the natural language descriptions of coding tasks that would cause the LLMs to produce incorrect outputs. We have formally proved the theoretical properties of the method and also conducted extensive experiments to evaluate the robustness of four state-of-the-art art LLMs: Gemini-pro, Codex, Llamma2 and Falcon 7B, and have found that we are able to rank these with confidence from best to worst. Moreover, we have also studied how robustness varies in different scenarios, including the variations with the topic of the coding task and with the complexity of its sample solution, and found that robustness is lower for more complex tasks and also lower for more advanced topics, such as multi-threading and data structures. --- ### Rethinking Preference Alignment for Diffusion Models with Classifier-Free Guidance Authors: Zhou Jiang, Yandong Wen, Zhen Liu Published: 2026-02-21 Categories: cs.CV arXiv: https://arxiv.org/abs/2602.18799v1 Aligning large-scale text-to-image diffusion models with nuanced human preferences remains challenging. While direct preference optimization (DPO) is simple and effective, large-scale finetuning often shows a generalization gap. We take inspiration from test-time guidance and cast preference alignment as classifier-free guidance (CFG): a finetuned preference model acts as an external control signal during sampling. Building on this view, we propose a simple method that improves alignment without retraining the base model. To further enhance generalization, we decouple preference learning into two modules trained on positive and negative data, respectively, and form a \emph{contrastive guidance} vector at inference by subtracting their predictions (positive minus negative), scaled by a user-chosen strength and added to the base prediction at each step. This yields a sharper and controllable alignment signal. We evaluate on Stable Diffusion 1.5 and Stable Diffusion XL with Pick-a-Pic v2 and HPDv3, showing consistent quantitative and qualitative gains. --- ### Carbon-aware decentralized dynamic task offloading in MIMO-MEC networks via multi-agent reinforcement learning Authors: Mubshra Zulfiqar, Muhammad Ayzed Mirza, Basit Qureshi Published: 2026-02-21 Categories: cs.DC, cs.AI, cs.LG arXiv: https://arxiv.org/abs/2602.18797v1 Massive internet of things microservices require integrating renewable energy harvesting into mobile edge computing (MEC) for sustainable eScience infrastructures. Spatiotemporal mismatches between stochastic task arrivals and intermittent green energy along with complex inter-user interference in multi-antenna (MIMO) uplinks complicate real-time resource management. Traditional centralized optimization and off-policy reinforcement learning struggle with scalability and signaling overhead in dense networks. This paper proposes CADDTO-PPO, a carbon-aware decentralized dynamic task offloading framework based on multi-agent proximal policy optimization. The multi-user MIMO-MEC system is modeled as a Decentralized Partially Observable Markov Decision Process (DEC-POMDP) to jointly minimize carbon emissions and buffer latency and energy wastage. A scalable architecture utilizes decentralized execution with parameter sharing (DEPS), which enables autonomous IoT agents to make fine-grained power control and offloading decisions based solely on local observations. Additionally, a carbon-first reward structure adaptively prioritizes green time slots for data transmission to decouple system throughput from grid-dependent carbon footprints. Finally, experimental results demonstrate CADDTO-PPO outperforms deep deterministic policy gradient (DDPG) and lyapunov-based baselines. The framework achieves the lowest carbon intensity and maintains near-zero packet overflow rates under extreme traffic loads. Architectural profiling validates the framework to demonstrate a constant $O(1)$ inference complexity and theoretical lightweight feasibility for future generation sustainable IoT deployments. --- ### Vectorized Bayesian Inference for Latent Dirichlet-Tree Allocation Authors: Zheng Wang, Nizar Bouguila Published: 2026-02-21 Categories: cs.LG, stat.ML arXiv: https://arxiv.org/abs/2602.18795v1 Latent Dirichlet Allocation (LDA) is a foundational model for discovering latent thematic structure in discrete data, but its Dirichlet prior cannot represent the rich correlations and hierarchical relationships often present among topics. We introduce the framework of Latent Dirichlet-Tree Allocation (LDTA), a generalization of LDA that replaces the Dirichlet prior with an arbitrary Dirichlet-Tree (DT) distribution. LDTA preserves LDA's generative structure but enables expressive, tree-structured priors over topic proportions. To perform inference, we develop universal mean-field variational inference and Expectation Propagation, providing tractable updates for all DT. We reveal the vectorized nature of the two inference methods through theoretical development, and perform fully vectorized, GPU-accelerated implementations. The resulting framework substantially expands the modeling capacity of LDA while maintaining scalability and computational efficiency. --- ### From Few-Shot to Zero-Shot: Towards Generalist Graph Anomaly Detection Authors: Yixin Liu, Shiyuan Li, Yu Zheng, Qingfeng Chen, Chengqi Zhang, Philip S. Yu, Shirui Pan Published: 2026-02-21 Categories: cs.LG arXiv: https://arxiv.org/abs/2602.18793v1 Graph anomaly detection (GAD) is critical for identifying abnormal nodes in graph-structured data from diverse domains, including cybersecurity and social networks. The existing GAD methods often focus on the learning paradigms of "one-model-for-one-dataset", requiring dataset-specific training for each dataset to achieve optimal performance. However, this paradigm suffers from significant limitations, such as high computational and data costs, limited generalization and transferability to new datasets, and challenges in privacy-sensitive scenarios where access to full datasets or sufficient labels is restricted. To address these limitations, we propose a novel generalist GAD paradigm that aims to develop a unified model capable of detecting anomalies on multiple unseen datasets without extensive retraining/fine-tuning or dataset-specific customization. To this end, we propose ARC, a few-shot generalist GAD method that leverages in-context learning and requires only a few labeled normal samples at inference time. Specifically, ARC consists of three core modules: a feature Alignment module to unify and align features across datasets, a Residual GNN encoder to capture dataset-agnostic anomaly representations, and a cross-attentive in-Context learning module to score anomalies using few-shot normal context. Building on ARC, we further introduce ARC_zero for the zero-shot generalist GAD setting, which selects representative pseudo-normal nodes via a pseudo-context mechanism and thus enables fully label-free inference on unseen datasets. Extensive experiments on 17 real-world graph datasets demonstrate that both ARC and ARC_zero effectively detect anomalies, exhibit strong generalization ability, and perform efficiently under few-shot and zero-shot settings. --- ### MaskDiME: Adaptive Masked Diffusion for Precise and Efficient Visual Counterfactual Explanations Authors: Changlu Guo, Anders Nymark Christensen, Anders Bjorholm Dahl, Morten Rieger Hannemose Published: 2026-02-21 Categories: cs.CV arXiv: https://arxiv.org/abs/2602.18792v3 Visual counterfactual explanations aim to reveal the minimal semantic modifications that can alter a model's prediction, providing causal and interpretable insights into deep neural networks. However, existing diffusion-based counterfactual generation methods are often computationally expensive, slow to sample, and imprecise in localizing the modified regions. To address these limitations, we propose MaskDiME, a simple, fast, yet effective diffusion framework that unifies semantic consistency and spatial precision through localized sampling. Our approach adaptively focuses on decision-relevant regions to achieve localized and semantically consistent counterfactual generation while preserving high image fidelity. Our training-free framework, MaskDiME, performs inference over 30x faster than the baseline and achieves comparable or state-of-the-art performance across five benchmark datasets spanning diverse visual domains, establishing a practical and generalizable solution for efficient counterfactual explanation. --- ### BURMESE-SAN: Burmese NLP Benchmark for Evaluating Large Language Models Authors: Thura Aung, Jann Railey Montalan, Jian Gang Ngui, Peerat Limkonchotiwat Published: 2026-02-21 Categories: cs.CL arXiv: https://arxiv.org/abs/2602.18788v3 We introduce BURMESE-SAN, the first holistic benchmark that systematically evaluates large language models (LLMs) for Burmese across three core NLP competencies: understanding (NLU), reasoning (NLR), and generation (NLG). BURMESE-SAN consolidates seven subtasks spanning these competencies, including Question Answering, Sentiment Analysis, Toxicity Detection, Causal Reasoning, Natural Language Inference, Abstractive Summarization, and Machine Translation, several of which were previously unavailable for Burmese. The benchmark is constructed through a rigorous native-speaker-driven process to ensure linguistic naturalness, fluency, and cultural authenticity while minimizing translation-induced artifacts. We conduct a large-scale evaluation of both open-weight and commercial LLMs to examine challenges in Burmese modeling arising from limited pretraining coverage, rich morphology, and syntactic variation. Our results show that Burmese performance depends more on architectural design, language representation, and instruction tuning than on model scale alone. In particular, Southeast Asia regional fine-tuning and newer model generations yield substantial gains. Finally, we release BURMESE-SAN as a public leaderboard to support systematic evaluation and sustained progress in Burmese and other low-resource languages. https://leaderboard.sea-lion.ai/detailed/MY --- ### CaliCausalRank: Calibrated Multi-Objective Ad Ranking with Robust Counterfactual Utility Optimization Authors: Xikai Yang, Sebastian Sun, Yilin Li, Yue Xing, Ming Wang, Yang Wang Published: 2026-02-21 Categories: cs.LG, cs.IR arXiv: https://arxiv.org/abs/2602.18786v1 Ad ranking systems must simultaneously optimize multiple objectives including click-through rate (CTR), conversion rate (CVR), revenue, and user experience metrics. However, production systems face critical challenges: score scale inconsistency across traffic segments undermines threshold transferability, and position bias in click logs causes offline-online metric discrepancies. We propose CaliCausalRank, a unified framework that integrates training-time scale calibration, constraint-based multi-objective optimization, and robust counterfactual utility estimation. Our approach treats score calibration as a first-class training objective rather than post-hoc processing, employs Lagrangian relaxation for constraint satisfaction, and utilizes variance-reduced counterfactual estimators for reliable offline evaluation. Experiments on the Criteo and Avazu datasets demonstrate that CaliCausalRank achieves 1.1% relative AUC improvement, 31.6% calibration error reduction, and 3.2% utility gain compared to the best baseline (PairRank) while maintaining consistent performance across different traffic segments. --- ### MANATEE: Inference-Time Lightweight Diffusion Based Safety Defense for LLMs Authors: Chun Yan Ryan Kan, Tommy Tran, Vedant Yadav, Ava Cai, Kevin Zhu, Ruizhe Li, Maheep Chaudhary Published: 2026-02-21 Categories: cs.CR, cs.AI, cs.CL, cs.LG arXiv: https://arxiv.org/abs/2602.18782v1 Defending LLMs against adversarial jailbreak attacks remains an open challenge. Existing defenses rely on binary classifiers that fail when adversarial input falls outside the learned decision boundary, and repeated fine-tuning is computationally expensive while potentially degrading model capabilities. We propose MANATEE, an inference-time defense that uses density estimation over a benign representation manifold. MANATEE learns the score function of benign hidden states and uses diffusion to project anomalous representations toward safe regions--requiring no harmful training data and no architectural modifications. Experiments across Mistral-7B-Instruct, Llama-3.1-8B-Instruct, and Gemma-2-9B-it demonstrate that MANATEE reduce Attack Success Rate by up to 100\% on certain datasets, while preserving model utility on benign inputs. --- ### ArabicNumBench: Evaluating Arabic Number Reading in Large Language Models Authors: Anas Alhumud, Abdulaziz Alhammadi, Muhammad Badruddin Khan Published: 2026-02-21 Categories: cs.CL, cs.AI arXiv: https://arxiv.org/abs/2602.18776v1 We present ArabicNumBench, a comprehensive benchmark for evaluating large language models on Arabic number reading tasks across Eastern Arabic-Indic numerals (0-9 in Arabic script) and Western Arabic numerals (0-9). We evaluate 71 models from 10 providers using four prompting strategies (zero-shot, zero-shot CoT, few-shot, few-shot CoT) on 210 number reading tasks spanning six contextual categories: pure numerals, addresses, dates, quantities, and prices. Our evaluation comprises 59,010 individual test cases and tracks extraction methods to measure structured output generation. Evaluation reveals substantial performance variation, with accuracy ranging from 14.29\% to 99.05\% across models and strategies. Few-shot Chain-of-Thought prompting achieves 2.8x higher accuracy than zero-shot approaches (80.06\% vs 28.76\%). A striking finding emerges: models achieving elite accuracy (98-99\%) often produce predominantly unstructured output, with most responses lacking Arabic CoT markers. Only 6 models consistently generate structured output across all test cases, while the majority require fallback extraction methods despite high numerical accuracy. Comprehensive evaluation of 281 model-strategy combinations demonstrates that numerical accuracy and instruction-following represent distinct capabilities, establishing baselines for Arabic number comprehension and providing actionable guidance for model selection in production Arabic NLP systems. --- ### LAMMI-Pathology: A Tool-Centric Bottom-Up LVLM-Agent Framework for Molecularly Informed Medical Intelligence in Pathology Authors: Haoyang Su, Shaoting Zhang, Xiaosong Wang Published: 2026-02-21 Categories: cs.AI arXiv: https://arxiv.org/abs/2602.18773v1 The emergence of tool-calling-based agent systems introduces a more evidence-driven paradigm for pathology image analysis in contrast to the coarse-grained text-image diagnostic approaches. With the recent large-scale experimental adoption of spatial transcriptomics technologies, molecularly validated pathological diagnosis is becoming increasingly open and accessible. In this work, we propose LAMMI-Pathology (LVLM-Agent System for Molecularly Informed Medical Intelligence in Pathology), a scalable agent framework for domain-specific agent tool-calling. LAMMI-Pathology adopts a tool-centric, bottom-up architecture in which customized domain-adaptive tools serve as the foundation. These tools are clustered by domain style to form component agents, which are then coordinated through a top-level planner hierarchically, avoiding excessively long context lengths that could induce task drift. Based on that, we introduce a novel trajectory construction mechanism based on Atomic Execution Nodes (AENs), which serve as reliable and composable units for building semi-simulated reasoning trajectories that capture credible agent-tool interactions. Building on this foundation, we develop a trajectory-aware fine-tuning strategy that aligns the planner's decision-making process with these multi-step reasoning trajectories, thereby enhancing inference robustness in pathology understanding and its adaptive use of the customized toolset. --- ### GLaDiGAtor: Language-Model-Augmented Multi-Relation Graph Learning for Predicting Disease-Gene Associations Authors: Osman Onur Kuzucu, Tunca Doğan Published: 2026-02-21 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2602.18769v1 Understanding disease-gene associations is essential for unravelling disease mechanisms and advancing diagnostics and therapeutics. Traditional approaches based on manual curation and literature review are labour-intensive and not scalable, prompting the use of machine learning on large biomedical data. In particular, graph neural networks (GNNs) have shown promise for modelling complex biological relationships. To address limitations in existing models, we propose GLaDiGAtor (Graph Learning-bAsed DIsease-Gene AssociaTiOn pRediction), a novel GNN framework with an encoder-decoder architecture for disease-gene association prediction. GLaDiGAtor constructs a heterogeneous biological graph integrating gene-gene, disease-disease, and gene-disease interactions from curated databases, and enriches each node with contextual features from well-known language models (ProtT5 for protein sequences and BioBERT for disease text). In evaluations, our model achieves superior predictive accuracy and generalisation, outperforming 14 existing methods. Literature-supported case studies confirm the biological relevance of high-confidence novel predictions, highlighting GLaDiGAtor's potential to discover candidate disease genes. These results underscore the power of graph convolutional networks in biomedical informatics and may ultimately facilitate drug discovery by revealing new gene-disease links. The source code and processed datasets are publicly available at https://github.com/HUBioDataLab/GLaDiGAtor. --- ### Nazrin: An Atomic Neural Proof Automation Tactic in Lean 4 Authors: Leni Aniva, Iori Oikawa, David Dill, Clark Barrett Published: 2026-02-21 Categories: cs.LO, cs.LG arXiv: https://arxiv.org/abs/2602.18767v3 In Machine-Assisted Theorem Proving, a theorem proving agent searches for a sequence of expressions and tactics that can prove a statement in a proof assistant. In this work, we introduce several novel concepts and capabilities to address obstacles faced by machine-assisted theorem proving. We first present a set of \textbf{atomic tactics}, a small finite set of tactics capable of proving any provable statement in Lean. We then introduce a \textbf{transposing atomization} algorithm which turns arbitrary proof expressions into a series of atomic tactics. We next introduce the \textbf{ExprGraph} data structure, which provides a succinct representation for Lean expressions. Finally, we present the \textbf{Nazrin Prover}, short for \textbf{N}eural \textbf{A}tomi\textbf{z}e\textbf{r} for \textbf{In}habitation Problems, a graph neural network-based theorem proving agent using atomic tactics and ExprGraph. Nazrin circumvents many challenges faced by existing proving agents by exclusively dispatching atomic tactics, and it is robust enough to both train and evaluate on consumer-grade hardware. We demonstrate the potential of tools like Nazrin using theorems from Lean's standard library and from Mathlib. --- ### Initialization matters in few-shot adaptation of vision-language models for histopathological image classification Authors: Pablo Meseguer, Rocío del Amor, Valery Naranjo Published: 2026-02-21 Categories: cs.CV arXiv: https://arxiv.org/abs/2602.18766v1 Vision language models (VLM) pre-trained on datasets of histopathological image-caption pairs enabled zero-shot slide-level classification. The ability of VLM image encoders to extract discriminative features also opens the door for supervised fine-tuning for whole-slide image (WSI) classification, ideally using few labeled samples. Slide-level prediction frameworks require the incorporation of multiple instance learning (MIL) due to the gigapixel size of the WSI. Following patch-level feature extraction and aggregation, MIL frameworks rely on linear classifiers trained on top of the slide-level aggregated features. Classifier weight initialization has a large influence on Linear Probing performance in efficient transfer learning (ETL) approaches based on few-shot learning. In this work, we propose Zero-Shot Multiple-Instance Learning (ZS-MIL) to address the limitations of random classifier initialization that underperform zero-shot prediction in MIL problems. ZS-MIL uses the class-level embeddings of the VLM text encoder as the classification layer's starting point to compute each sample's bag-level probabilities. Through multiple experiments, we demonstrate the robustness of ZS-MIL compared to well-known weight initialization techniques both in terms of performance and variability in an ETL few-shot scenario for subtyping prediction. --- ### A high-resolution nationwide urban village mapping product for 342 Chinese cities based on foundation models Authors: Lubin Bai, Sheng Xiao, Ziyu Yin, Haoyu Wang, Siyang Wu, Xiuyuan Zhang, Shihong Du Published: 2026-02-21 Categories: cs.CV arXiv: https://arxiv.org/abs/2602.18765v1 Urban Villages (UVs) represent a distinctive form of high-density informal settlement embedded within China's rapidly urbanizing cities. Accurate identification of UVs is critical for urban governance, renewal, and sustainable development. But due to the pronounced heterogeneity and diversity of UVs across China's vast territory, a consistent and reliable nationwide dataset has been lacking. In this work, we present GeoLink-UV, a high-resolution nationwide UV mapping product that clearly delineates the locations and boundaries of UVs in 342 Chinese cities. The dataset is derived from multisource geospatial data, including optical remote sensing images and geo-vector data, and is generated through a foundation model-driven mapping framework designed to address the generalization issues and improve the product quality. A geographically stratified accuracy assessment based on independent samples from 28 cities confirms the reliability and scientific credibility of the nationwide dataset across heterogeneous urban contexts. Based on this nationwide product, we reveal substantial interregional disparities in UV prevalence and spatial configuration. On average, UV areas account for 8 % of built-up land, with marked clustering in central and south China. Building-level analysis further confirms a consistent low-rise, high-density development pattern of UVs nationwide, while highlighting regionally differentiated morphological characteristics. The GeoLink-UV dataset provides an open and systematically validated geospatial foundation for urban studies, informal settlement monitoring, and evidence-based urban renewal planning, and contributes directly to large-scale assessments aligned with Sustainable Development Goal 11. The GeoLink-UV dataset introduced in this article is freely available at https://doi.org/10.5281/zenodo.18688062. --- ### The Convergence of Schema-Guided Dialogue Systems and the Model Context Protocol Authors: Andreas Schlapbach Published: 2026-02-21 Categories: cs.AI, cs.CL arXiv: https://arxiv.org/abs/2602.18764v2 This paper establishes a fundamental convergence: Schema-Guided Dialogue (SGD) and the Model Context Protocol (MCP) represent two manifestations of a unified paradigm for deterministic, auditable LLM-agent interaction. SGD, designed for dialogue-based API discovery (2019), and MCP, now the de facto standard for LLM-tool integration, share the same core insight -- that schemas can encode not just tool signatures but operational constraints and reasoning guidance. By analyzing this convergence, we extract five foundational principles for schema design: (1) Semantic Completeness over Syntactic Precision, (2) Explicit Action Boundaries, (3) Failure Mode Documentation, (4) Progressive Disclosure Compatibility, and (5) Inter-Tool Relationship Declaration. These principles reveal three novel insights: first, SGD's original design was fundamentally sound and should be inherited by MCP; second, both frameworks leave failure modes and inter-tool relationships unexploited -- gaps we identify and resolve; third, progressive disclosure emerges as a critical production-scaling insight under real-world token constraints. We provide concrete design patterns for each principle. These principles position schema-driven governance as a scalable mechanism for AI system oversight without requiring proprietary system inspection -- central to Software 3.0. --- ### TAG: Thinking with Action Unit Grounding for Facial Expression Recognition Authors: Haobo Lin, Tianyi Bai, Jiajun Zhang, Xuanhao Chang, Sheng Lu, Fangming Gu, Zengjie Hu, Wentao Zhang Published: 2026-02-21 Categories: cs.CV, cs.AI arXiv: https://arxiv.org/abs/2602.18763v1 Facial Expression Recognition (FER) is a fine-grained visual understanding task where reliable predictions require reasoning over localized and meaningful facial cues. Recent vision--language models (VLMs) enable natural language explanations for FER, but their reasoning is often ungrounded, producing fluent yet unverifiable rationales that are weakly tied to visual evidence and prone to hallucination, leading to poor robustness across different datasets. We propose TAG (Thinking with Action Unit Grounding), a vision--language framework that explicitly constrains multimodal reasoning to be supported by facial Action Units (AUs). TAG requires intermediate reasoning steps to be grounded in AU-related facial regions, yielding predictions accompanied by verifiable visual evidence. The model is trained via supervised fine-tuning on AU-grounded reasoning traces followed by reinforcement learning with an AU-aware reward that aligns predicted regions with external AU detectors. Evaluated on RAF-DB, FERPlus, and AffectNet, TAG consistently outperforms strong open-source and closed-source VLM baselines while simultaneously improving visual faithfulness. Ablation and preference studies further show that AU-grounded rewards stabilize reasoning and mitigate hallucination, demonstrating the importance of structured grounded intermediate representations for trustworthy multimodal reasoning in FER. The code will be available at https://github.com/would1920/FER_TAG . --- ### Bounds and Identification of Joint Probabilities of Potential Outcomes and Observed Variables under Monotonicity Assumptions Authors: Naoya Hashimoto, Yuta Kawakami, Jin Tian Published: 2026-02-21 Categories: stat.ML, cs.LG arXiv: https://arxiv.org/abs/2602.18762v1 Evaluating joint probabilities of potential outcomes and observed variables, and their linear combinations, is a fundamental challenge in causal inference. This paper addresses the bounding and identification of these probabilities in settings with discrete treatment and discrete ordinal outcome. We propose new families of monotonicity assumptions and formulate the bounding problem as a linear programming problem. We further introduce a new monotonicity assumption specifically to achieve identification. Finally, we present numerical experiments to validate our methods and demonstrate their application using real-world datasets. --- ### Towards Reliable Negative Sampling for Recommendation with Implicit Feedback via In-Community Popularity Authors: Chen Chen, Haobo Lin, Yuanbo Xu Published: 2026-02-21 Categories: cs.IR, cs.AI arXiv: https://arxiv.org/abs/2602.18759v1 Learning from implicit feedback is a fundamental problem in modern recommender systems, where only positive interactions are observed and explicit negative signals are unavailable. In such settings, negative sampling plays a critical role in model training by constructing negative items that enable effective preference learning and ranking optimization. However, designing reliable negative sampling strategies remains challenging, as they must simultaneously ensure realness, hardness, and interpretability. To this end, we propose \textbf{ICPNS (In-Community Popularity Negative Sampling)}, a novel framework that leverages user community structure to identify reliable and informative negative samples. Our approach is grounded in the insight that item exposure is driven by latent user communities. By identifying these communities and utilizing in-community popularity, ICPNS effectively approximates the probability of item exposure. Consequently, items that are popular within a user's community but remain unclicked are identified as more reliable true negatives. Extensive experiments on four benchmark datasets demonstrate that ICPNS yields consistent improvements on graph-based recommenders and competitive performance on MF-based models, outperforming representative negative sampling strategies under a unified evaluation protocol. --- ### UFO: Unlocking Ultra-Efficient Quantized Private Inference with Protocol and Algorithm Co-Optimization Authors: Wenxuan Zeng, Chao Yang, Tianshi Xu, Bo Zhang, Changrui Ren, Jin Dong, Meng Li Published: 2026-02-21 Categories: cs.CR, cs.AI arXiv: https://arxiv.org/abs/2602.18758v1 Private convolutional neural network (CNN) inference based on secure two-party computation (2PC) suffers from high communication and latency overhead, especially from convolution layers. In this paper, we propose UFO, a quantized 2PC inference framework that jointly optimizes the 2PC protocols and quantization algorithm. UFO features a novel 2PC protocol that systematically combines the efficient Winograd convolution algorithm with quantization to improve inference efficiency. However, we observe that naively combining quantization and Winograd convolution faces the following challenges: 1) From the inference perspective, Winograd transformations introduce extensive additions and require frequent bit width conversions to avoid inference overflow, leading to non-negligible communication overhead; 2) From the training perspective, Winograd transformations introduce weight outliers that make quantization-aware training (QAT) difficult, resulting in inferior model accuracy. To address these challenges, we co-optimize both protocol and algorithm. 1) At the protocol level, we propose a series of graph-level optimizations for 2PC inference to minimize the communication. 2) At the algorithm level, we develop a mixed-precision QAT algorithm based on layer sensitivity to optimize model accuracy given communication constraints. To accommodate the outliers, we further introduce a 2PC-friendly bit re-weighting algorithm to increase the representation range without explicitly increasing bit widths. With extensive experiments, UFO demonstrates 11.7x, 3.6x, and 6.3x communication reduction with 1.29%, 1.16%, and 1.29% higher accuracy compared to state-of-the-art frameworks SiRNN, COINN, and CoPriv, respectively. --- ### Driving Like Yourself: A Benchmark for Closed-Loop Personalized End-to-End Autonomous Driving Authors: Xiaoru Dong, Ruiqin Li, Xiao Han, Zhenxuan Wu, Jiamin Wang, Jian Chen, Qi Jiang, SM Yiu, Xinge Zhu, Yuexin Ma Published: 2026-02-21 Categories: cs.CV arXiv: https://arxiv.org/abs/2602.18757v3 Human driving behavior is inherently diverse, yet most end-to-end autonomous driving (E2E-AD) systems learn a single average driving style, neglecting individual differences. Achieving personalized E2E-AD faces challenges across three levels: limited real-world datasets with individual-level annotations, a lack of quantitative metrics for evaluating personal driving styles, and the absence of algorithms that can learn stylized representations from users' trajectories. To address these gaps, we propose Person2Drive, a comprehensive personalized E2E-AD platform and benchmark. It includes an open-source, flexible data collection system that simulates realistic scenarios to generate scalable, diverse personalized driving datasets; style vector-based evaluation metrics with Maximum Mean Discrepancy and KL divergence to comprehensively quantify individual driving behaviors; and a personalized E2E-AD framework with a style reward model that efficiently adapts E2E models for safe and individualized driving. Crucially, our framework enables plug-and-play personalization by fine-tuning only the trajectory prediction head, preserving the pretrained base model and ensuring safety. Extensive experiments demonstrate that Person2Drive enables fine-grained analysis and effective personalization, while preserving driving performance and success rate even in challenging scenarios. --- ### Optimizing ID Consistency in Multimodal Large Models: Facial Restoration via Alignment, Entanglement, and Disentanglement Authors: Yuran Dong, Hang Dai, Mang Ye Published: 2026-02-21 Categories: cs.CV, cs.GR arXiv: https://arxiv.org/abs/2602.18752v1 Multimodal editing large models have demonstrated powerful editing capabilities across diverse tasks. However, a persistent and long-standing limitation is the decline in facial identity (ID) consistency during realistic portrait editing. Due to the human eye's high sensitivity to facial features, such inconsistency significantly hinders the practical deployment of these models. Current facial ID preservation methods struggle to achieve consistent restoration of both facial identity and edited element IP due to Cross-source Distribution Bias and Cross-source Feature Contamination. To address these issues, we propose EditedID, an Alignment-Disentanglement-Entanglement framework for robust identity-specific facial restoration. By systematically analyzing diffusion trajectories, sampler behaviors, and attention properties, we introduce three key components: 1) Adaptive mixing strategy that aligns cross-source latent representations throughout the diffusion process. 2) Hybrid solver that disentangles source-specific identity attributes and details. 3) Attentional gating mechanism that selectively entangles visual elements. Extensive experiments show that EditedID achieves state-of-the-art performance in preserving original facial ID and edited element IP consistency. As a training-free and plug-and-play solution, it establishes a new benchmark for practical and reliable single/multi-person facial identity restoration in open-world settings, paving the way for the deployment of multimodal editing large models in real-person editing scenarios. The code is available at https://github.com/NDYBSNDY/EditedID. --- ### Federated Reasoning Distillation Framework with Model Learnability-Aware Data Allocation Authors: Wei Guo, Siyuan Lu, Xiangdong Ran, Yiqi Tong, Yikun Ban, Zelong Xu, Jing Fan, Zixuan Huang, Xiao Zhang, Zhaojun Hu, Fuzhen Zhuang Published: 2026-02-21 Categories: cs.AI arXiv: https://arxiv.org/abs/2602.18749v1 Data allocation plays a critical role in federated large language model (LLM) and small language models (SLMs) reasoning collaboration. Nevertheless, existing data allocation methods fail to address an under-explored challenge in collaboration: bidirectional model learnability gap, where client-side SLMs cannot identify high-reward samples matching their learnability constraints for effective knowledge transfer from LLMs, while LLMs struggle to select samples contributing novel knowledge beyond their existing data. Furthermore, these collaboration frameworks face another key challenge: domain-agnostic reasoning transfer, where existing reasoning transfer methods fail to flexibly adapt to the local domain data, preventing SLMs from effectively acquiring step-by-step reasoning abilities within from general LLM. To address these challenges, we propose LaDa, a federated reasoning distillation framework with model learnability-aware data allocation. It introduces a model learnability-aware data filter that adaptively allocates high-reward samples based on the learnability gap between each SLM and LLM pair, effectively facilitating bidirectional knowledge transfer. We further design a domain adaptive reasoning distillation method that aligns joint probabilities of reasoning paths on filtered high-reward samples through contrastive distillation learning between SLM and LLM, enabling SLM to capture underlying reasoning patterns under local data distribution. LaDa operates as a plug-in module for existing collaboration frameworks, adapting knowledge transfer based on model learnability gaps. --- ### Benchmarking Computational Pathology Foundation Models For Semantic Segmentation Authors: Lavish Ramchandani, Aashay Tinaikar, Dev Kumar Das, Rohit Garg, Tijo Thomas Published: 2026-02-21 Categories: cs.CV arXiv: https://arxiv.org/abs/2602.18747v1 In recent years, foundation models such as CLIP, DINO,and CONCH have demonstrated remarkable domain generalization and unsupervised feature extraction capabilities across diverse imaging tasks. However, systematic and independent evaluations of these models for pixel-level semantic segmentation in histopathology remain scarce. In this study, we propose a robust benchmarking approach to asses 10 foundational models on four histopathological datasets covering both morphological tissue-region and cellular/nuclear segmentation tasks. Our method leverages attention maps of foundation models as pixel-wise features, which are then classified using a machine learning algorithm, XGBoost, enabling fast, interpretable, and model-agnostic evaluation without finetuning. We show that the vision language foundation model, CONCH performed the best across datasets when compared to vision-only foundation models, with PathDino as close second. Further analysis shows that models trained on distinct histopathology cohorts capture complementary morphological representations, and concatenating their features yields superior segmentation performance. Concatenating features from CONCH, PathDino and CellViT outperformed individual models across all the datasets by 7.95% (averaged across the datasets), suggesting that ensembles of foundation models can better generalize to diverse histopathological segmentation tasks. --- ### Bridging Modality Disconnect in Self-Reflection via Closed-Loop Visually Grounded Verification Authors: Haoyu Zhang, Yuwei Wu, Pengxiang Li, Xintong Zhang, Zhi Gao, Rui Gao, Mingyang Gao, Che Sun, Yunde Jia Published: 2026-02-21 Categories: cs.CV arXiv: https://arxiv.org/abs/2602.18746v3 In the era of Vision-Language Models (VLMs), enhancing multimodal reasoning capabilities remains a critical challenge, particularly in handling ambiguous or complex visual inputs, where initial inferences often lead to hallucinations or logic errors. Existing VLMs often produce plausible yet ungrounded answers, and even when prompted to "reflect", their corrections may remain detached from the image evidence. To address this, we propose the MIRROR framework for Multimodal Iterative Reasoning via Reflection On visual Regions. By embedding visual reflection as a core mechanism, MIRROR is formulated as a closed-loop process comprising draft, critique, region-based verification, and revision, which are repeated until the output is visually grounded. To facilitate training of this model, we construct **ReflectV**, a visual reflective dataset for multi-turn supervision that explicitly contains reflection triggers, region-based verification actions, and answer revision grounded in visual evidence. Experiments on both general vision-language benchmarks and representative vision-language reasoning benchmarks show that MIRROR improves correctness and reduces visual hallucinations, demonstrating the value of training reflection as an evidence-seeking, region-aware verification process rather than a purely textual revision step. --- ### Synthesizing Multimodal Geometry Datasets from Scratch and Enabling Visual Alignment via Plotting Code Authors: Haobo Lin, Tianyi Bai, Chen Chen, Jiajun Zhang, Bohan Zeng, Wentao Zhang, Binhang Yuan Published: 2026-02-21 Categories: cs.CV, cs.AI arXiv: https://arxiv.org/abs/2602.18745v1 Multimodal geometry reasoning requires models to jointly understand visual diagrams and perform structured symbolic inference, yet current vision--language models struggle with complex geometric constructions due to limited training data and weak visual--symbolic alignment. We propose a pipeline for synthesizing complex multimodal geometry problems from scratch and construct a dataset named \textbf{GeoCode}, which decouples problem generation into symbolic seed construction, grounded instantiation with verification, and code-based diagram rendering, ensuring consistency across structure, text, reasoning, and images. Leveraging the plotting code provided in GeoCode, we further introduce code prediction as an explicit alignment objective, transforming visual understanding into a supervised structured prediction task. GeoCode exhibits substantially higher structural complexity and reasoning difficulty than existing benchmarks, while maintaining mathematical correctness through multi-stage validation. Extensive experiments show that models trained on GeoCode achieve consistent improvements on multiple geometry benchmarks, demonstrating both the effectiveness of the dataset and the proposed alignment strategy. The code will be available at https://github.com/would1920/GeoCode. --- ### RadioGen3D: 3D Radio Map Generation via Adversarial Learning on Large-Scale Synthetic Data Authors: Junshen Chen, Angzi Xu, Zezhong Zhang, Shiyao Zhang, Junting Chen, Shuguang Cui Published: 2026-02-21 Categories: cs.LG arXiv: https://arxiv.org/abs/2602.18744v1 Radio maps are essential for efficient radio resource management in future 6G and low-altitude networks. While deep learning (DL) techniques have emerged as an efficient alternative to conventional ray-tracing for radio map estimation (RME), most existing DL approaches are confined to 2D near-ground scenarios. They often fail to capture essential 3D signal propagation characteristics and antenna polarization effects, primarily due to the scarcity of 3D data and training challenges. To address these limitations, we present the RadioGen3D framework. First, we propose an efficient data synthesis method to generate high-quality 3D radio map data. By establishing a parametric target model that captures 2D ray-tracing and 3D channel fading characteristics, we derive realistic coefficient combinations from minimal real measurements, enabling the construction of a large-scale synthetic dataset, Radio3DMix. Utilizing this dataset, we propose a 3D model training scheme based on a conditional generative adversarial network (cGAN), yielding a 3D U-Net capable of accurate RME under diverse input feature combinations. Experimental results demonstrate that RadioGen3D surpasses all baselines in both estimation accuracy and speed. Furthermore, fine-tuning experiments verify its strong generalization capability via successful knowledge transfer. --- ### RoboCurate: Harnessing Diversity with Action-Verified Neural Trajectory for Robot Learning Authors: Seungku Kim, Suhyeok Jang, Byungjun Yoon, Dongyoung Kim, John Won, Jinwoo Shin Published: 2026-02-21 Categories: cs.RO, cs.AI, cs.CV arXiv: https://arxiv.org/abs/2602.18742v1 Synthetic data generated by video generative models has shown promise for robot learning as a scalable pipeline, but it often suffers from inconsistent action quality due to imperfectly generated videos. Recently, vision-language models (VLMs) have been leveraged to validate video quality, but they have limitations in distinguishing physically accurate videos and, even then, cannot directly evaluate the generated actions themselves. To tackle this issue, we introduce RoboCurate, a novel synthetic robot data generation framework that evaluates and filters the quality of annotated actions by comparing them with simulation replay. Specifically, RoboCurate replays the predicted actions in a simulator and assesses action quality by measuring the consistency of motion between the simulator rollout and the generated video. In addition, we unlock observation diversity beyond the available dataset via image-to-image editing and apply action-preserving video-to-video transfer to further augment appearance. We observe RoboCurate's generated data yield substantial relative improvements in success rates compared to using real data only, achieving +70.1% on GR-1 Tabletop (300 demos), +16.1% on DexMimicGen in the pre-training setup, and +179.9% in the challenging real-world ALLEX humanoid dexterous manipulation setting. --- ### Spectral Rendering Without a Spectral Renderer: Learned Spectral Codes for RGB Pipelines Authors: Jiaqi Yu, Dar'ya Guarnera, Giuseppe Claudio Guarnera Published: 2026-02-21 Categories: cs.GR, cs.CV arXiv: https://arxiv.org/abs/2602.18741v3 Spectral rendering reproduces the wavelength-dependent appearance that RGB rendering cannot: metamerism, colour shifts under spectrally rich illumination, coloured glass, metallic reflection and subsurface scattering. It requires a spectrally aware renderer, but the rendering ecosystem is predominantly RGB: game engines, cloud render farms, GPU rasterisers and most offline renderers consume RGB alone, with no access to the internal wavelength propagation that spectral rendering relies on. We introduce learned spectral codes, a compact linear representation that enables general spectral rendering through unmodified RGB pipelines. A trained encoder compresses any visible-range spectral quantity into a small latent vector; the codes are packed as RGB triplets and processed by an off-the-shelf RGB renderer. The codec preserves the arithmetic of light transport, exactly for scaling and addition and approximately for element-wise products, so two standard RGB passes suffice to produce results closely matching a full spectral rendering. The same codec handles reflectance, illumination, coloured glass, conductors, and participating media: one pipeline across material classes, with no per-material modification to the renderer. Where spectral asset data is unavailable, lightweight neural upsamplers bridge legacy RGB inputs to the latent space. Across diverse materials and illumination, the method yields substantial quality improvements on the cases where RGB rendering fails most visibly, at rendering cost comparable to efficient spectral-sampling techniques and with no changes to the renderer. --- ### Compact Hadamard Latent Codes for Efficient Spectral Rendering Authors: Jiaqi Yu, Dar'ya Guarnera, Giuseppe Claudio Guarnera Published: 2026-02-21 Categories: cs.GR, cs.CV arXiv: https://arxiv.org/abs/2602.18741v2 Spectral rendering accurately reproduces wavelength-dependent appearance but is computationally expensive, as shading must be evaluated at many wavelength samples and scales roughly linearly with the number of samples. It also requires spectral textures and lights throughout the rendering pipeline. We propose Hadamard spectral codes, a compact latent representation that enables spectral rendering using standard RGB rendering operations. Spectral images are approximated with a small number of RGB rendering passes, followed by a decoding step. Our key requirement is latent linearity: scaling and addition in spectral space correspond to scaling and addition of codes, and the element-wise product of spectra (for example reflectance times illumination) is approximated by the element-wise product of their latent codes. We show that an exact low-dimensional algebra-preserving representation cannot exist for arbitrary spectra when the latent dimension k is smaller than the number of spectral samples n. We therefore introduce a learned non-negative linear encoder and decoder architecture that preserves scaling and addition exactly while encouraging approximate multiplicativity under the Hadamard product. With k = 6, we render k/3 = 2 RGB images per frame using an unmodified RGB renderer, reconstruct the latent image, and decode to high-resolution spectra or XYZ or RGB. Experiments on 3D scenes demonstrate that k = 6 significantly reduces color error compared to RGB baselines while being substantially faster than naive n-sample spectral rendering. Using k = 9 provides higher-quality reference results. We further introduce a lightweight neural upsampling network that maps RGB assets directly to latent codes, enabling integration of legacy RGB content into the spectral pipeline while maintaining perceptually accurate colors in rendered images. --- ### HONEST-CAV: Hierarchical Optimization of Network Signals and Trajectories for Connected and Automated Vehicles with Multi-Agent Reinforcement Learning Authors: Ziyan Zhang, Changxin Wan, Peng Hao, Kanok Boriboonsomsin, Matthew J. Barth, Yongkang Liu, Seyhan Ucar, Guoyuan Wu Published: 2026-02-21 Categories: cs.LG, cs.AI, eess.SY arXiv: https://arxiv.org/abs/2602.18740v1 This study presents a hierarchical, network-level traffic flow control framework for mixed traffic consisting of Human-driven Vehicles (HVs), Connected and Automated Vehicles (CAVs). The framework jointly optimizes vehicle-level eco-driving behaviors and intersection-level traffic signal control to enhance overall network efficiency and decrease energy consumption. A decentralized Multi-Agent Reinforcement Learning (MARL) approach by Value Decomposition Network (VDN) manages cycle-based traffic signal control (TSC) at intersections, while an innovative Signal Phase and Timing (SPaT) prediction method integrates a Machine Learning-based Trajectory Planning Algorithm (MLTPA) to guide CAVs in executing Eco-Approach and Departure (EAD) maneuvers. The framework is evaluated across varying CAV proportions and powertrain types to assess its effects on mobility and energy performance. Experimental results conducted in a 4*4 real-world network demonstrate that the MARL-based TSC method outperforms the baseline model (i.e., Webster method) in speed, fuel consumption, and idling time. In addition, with MLTPA, HONEST-CAV benefits the traffic system further in energy consumption and idling time. With a 60% CAV proportion, vehicle average speed, fuel consumption, and idling time can be improved/saved by 7.67%, 10.23%, and 45.83% compared with the baseline. Furthermore, discussions on CAV proportions and powertrain types are conducted to quantify the performance of the proposed method with the impact of automation and electrification. --- ### When World Models Dream Wrong: Physical-Conditioned Adversarial Attacks against World Models Authors: Zhixiang Guo, Siyuan Liang, Andras Balogh, Noah Lunberry, Rong-Cheng Tu, Mark Jelasity, Dacheng Tao Published: 2026-02-21 Categories: cs.LG arXiv: https://arxiv.org/abs/2602.18739v1 Generative world models (WMs) are increasingly used to synthesize controllable, sensor-conditioned driving videos, yet their reliance on physical priors exposes novel attack surfaces. In this paper, we present Physical-Conditioned World Model Attack (PhysCond-WMA), the first white-box world model attack that perturbs physical-condition channels, such as HDMap embeddings and 3D-box features, to induce semantic, logic, or decision-level distortion while preserving perceptual fidelity. PhysCond-WMA is optimized in two stages: (1) a quality-preserving guidance stage that constrains reverse-diffusion loss below a calibrated threshold, and (2) a momentum-guided denoising stage that accumulates target-aligned gradients along the denoising trajectory for stable, temporally coherent semantic shifts. Extensive experimental results demonstrate that our approach remains effective while increasing FID by about 9% on average and FVD by about 3.9% on average. Under the targeted attack setting, the attack success rate (ASR) reaches 0.55. Downstream studies further show tangible risk, which using attacked videos for training decreases 3D detection performance by about 4%, and worsens open-loop planning performance by about 20%. These findings has for the first time revealed and quantified security vulnerabilities in generative world models, driving more comprehensive security checkers. --- ### LaS-Comp: Zero-shot 3D Completion with Latent-Spatial Consistency Authors: Weilong Yan, Haipeng Li, Hao Xu, Nianjin Ye, Yihao Ai, Shuaicheng Liu, Jingyu Hu Published: 2026-02-21 Categories: cs.CV, cs.RO arXiv: https://arxiv.org/abs/2602.18735v2 This paper introduces LaS-Comp, a zero-shot and category-agnostic approach that leverages the rich geometric priors of 3D foundation models to enable 3D shape completion across diverse types of partial observations. Our contributions are threefold: First, \ourname{} harnesses these powerful generative priors for completion through a complementary two-stage design: (i) an explicit replacement stage that preserves the partial observation geometry to ensure faithful completion; and (ii) an implicit refinement stage ensures seamless boundaries between the observed and synthesized regions. Second, our framework is training-free and compatible with different 3D foundation models. Third, we introduce Omni-Comp, a comprehensive benchmark combining real-world and synthetic data with diverse and challenging partial patterns, enabling a more thorough and realistic evaluation. Both quantitative and qualitative experiments demonstrate that our approach outperforms previous state-of-the-art approaches. Our code and data will be available at \href{https://github.com/DavidYan2001/LaS-Comp}{LaS-Comp}. --- ### Rethinking Retrieval-Augmented Generation as a Cooperative Decision-Making Problem Authors: Lichang Song, Ting Long, Yi Chang Published: 2026-02-21 Categories: cs.CL, cs.AI arXiv: https://arxiv.org/abs/2602.18734v2 Retrieval-Augmented Generation (RAG) has demonstrated strong effectiveness in knowledge-intensive tasks by grounding language generation in external evidence. Despite its success, many existing RAG systems are built based on a ranking-centric, asymmetric dependency paradigm, where the generation quality of the generator is highly dependent on reranking results of the reranker. To overcome this limitation, we propose Cooperative Retrieval-Augmented Generation (CoRAG), a framework that treats the reranker and the generator as peer decision-makers rather than being connected through an asymmetric dependency pipeline. By jointly optimizing their behaviors toward a shared task objective, the reranker and generator are encouraged to cooperate, ensuring that document reranking and generation work in concert to improve the final response. Experimental results demonstrate good generalization and improved generation stability of CoRAG, even when the model is trained on only around 10K PopQA samples. Our model released in https://github.com/CoderrrSong/CoRAG. --- ### A Prior-Aware Metric for Efficiently Distinguishing Memorization from Generalization in Large Language Models Authors: Trishita Tiwari, Ari Trachtenberg, G. Edward Suh Published: 2026-02-21 Categories: cs.LG arXiv: https://arxiv.org/abs/2602.18733v2 Training data leakage from Large Language Models (LLMs) raises serious concerns related to privacy, security, and copyright compliance. A central challenge in assessing this risk is distinguishing prefix-specific memorization of training data from the generation of statistically common sequences. Existing approaches to measuring memorization often conflate these phenomena, labeling outputs as memorized even when they arise from generalization over common patterns. Counterfactual memorization and other related metrics \citep{zhang2023counterfactual, wang2025generalization, lesci2024causal} provide principled solutions, however, their reliance on retraining multiple baseline models or parsing through the training data makes them computationally impractical at scale. This work introduces \emph{Prior-Aware memorization}, a theoretically grounded, lightweight and training-free criterion for identifying prefix-specific memorization in LLMs. The key idea is to evaluate whether a candidate suffix is strongly associated with its specific training prefix or whether it appears with high probability across many IID sampled sequences from the training data distribution due to statistical commonality. We correlate our metric with counterfactual memorization, and also evaluate it on the training corpora of two pre-trained models, LLaMA and OPT. Our results show that between 55\% and 90\% of sequences previously labeled as memorized fail our criterion and are consistent with statistical commonality --- ### Prior Aware Memorization: An Efficient Metric for Distinguishing Memorization from Generalization in Large Language Models Authors: Trishita Tiwari, Ari Trachtenberg, G. Edward Suh Published: 2026-02-21 Categories: cs.LG arXiv: https://arxiv.org/abs/2602.18733v1 Training data leakage from Large Language Models (LLMs) raises serious concerns related to privacy, security, and copyright compliance. A central challenge in assessing this risk is distinguishing genuine memorization of training data from the generation of statistically common sequences. Existing approaches to measuring memorization often conflate these phenomena, labeling outputs as memorized even when they arise from generalization over common patterns. Counterfactual Memorization provides a principled solution by comparing models trained with and without a target sequence, but its reliance on retraining multiple baseline models makes it computationally expensive and impractical at scale. This work introduces Prior-Aware Memorization, a theoretically grounded, lightweight and training-free criterion for identifying genuine memorization in LLMs. The key idea is to evaluate whether a candidate suffix is strongly associated with its specific training prefix or whether it appears with high probability across many unrelated prompts due to statistical commonality. We evaluate this metric on text from the training corpora of two pre-trained models, LLaMA and OPT, using both long sequences (to simulate copyright risks) and named entities (to simulate PII leakage). Our results show that between 55% and 90% of sequences previously labeled as memorized are in fact statistically common. Similar findings hold for the SATML training data extraction challenge dataset, where roughly 40% of sequences exhibit common-pattern behavior despite appearing only once in the training data. These results demonstrate that low frequency alone is insufficient evidence of memorization and highlight the importance of accounting for model priors when assessing leakage. ---