Current benchmarks for coding evaluate language models (LMs) on concrete, well-specified tasks such as fixing specific bugs or writing targeted tests. However, human programmers do not spend all day incessantly addressing isolated tasks. Instead, real-world software development is grounded in the pursuit of high-level goals, like improving user retention or reducing costs. Evaluating whether LMs can also iteratively develop code to better accomplish open-ended objectives without any explicit guidance remains an open challenge. To address this, we introduce CodeClash, a benchmark where LMs compete in multi-round tournaments to build the best codebase for achieving a competitive objective. Each round proceeds in two phases: agents edit their code, then their codebases compete head-to-head in a code arena that determines winners based on objectives like score maximization, resource acquisition, or survival. Whether it's writing notes, scrutinizing documentation, analyzing competition logs, or creating test suites, models must decide for themselves how to improve their codebases both absolutely and against their opponents. We run 1680 tournaments (25,200 rounds total) to evaluate 8 LMs across 6 arenas. Our results reveal that while models exhibit diverse development styles, they share fundamental limitations in strategic reasoning. Models also struggle with long-term codebase maintenance, as repositories become progressively messy and redundant. These limitations are stark: top models lose every round against expert human programmers. We open-source CodeClash to advance the study of autonomous, goal-oriented code development.

Though deep neural networks exhibit superior performance on various tasks, they are still plagued by adversarial examples. Adversarial training has been demonstrated to be the most effective method to defend against adversarial attacks. However, existing adversarial training methods show that the model robustness has apparent oscillations and overfitting issues in the training process, degrading the defense efficacy. To address these issues, we propose a novel framework called Parameter Interpolation Adversarial Training (PIAT). PIAT tunes the model parameters between each epoch by interpolating the parameters of the previous and current epochs. It makes the decision boundary of model change more moderate and alleviates the overfitting issue, helping the model converge better and achieving higher model robustness. In addition, we suggest using the Normalized Mean Square Error (NMSE) to further improve the robustness by aligning the relative magnitude of logits between clean and adversarial examples rather than the absolute magnitude. Extensive experiments conducted on several benchmark datasets demonstrate that our framework could prominently improve the robustness of both Convolutional Neural Networks (CNNs) and Vision Transformers (ViTs).

Vision Transformers have been tremendously successful in computer vision tasks. However, their large computational, memory, and energy demands are a challenge for edge inference on FPGAs -- a field that has seen a recent surge in demand. We recognize the benefits of recent works on logic and Look Up Table (LUT) based networks, such as LogicNets, NeuraLUT, DWN, among others, in offering models that simultaneously reduce both the memory and compute footprints. However, these models natively do not perform well on common vision tasks, such as CIFAR-10/100. In this work, we propose LL-ViT, a novel edge optimized vision transformer design that integrates layers of LUT neurons within the transformer architecture. Based on our characterization that reveals that a majority of model weights and computations are from the channel mixer (MLP layer), we design an alternate LUT-based channel mixer, and simultaneously develop an FPGA-based accelerator for LL-ViT. Contrary to some attempts to replace each multiplication with a table lookup, our architecture utilizes a neural learning approach which natively learns the LUT functions. This approach allows for reduced model sizes, and a computational and energy-efficient inference solution for vision transformer models. Evaluating on edge-suitable workloads, we achieve accuracies of 95.5% on CIFAR-10, 78.8% on CIFAR-100, and 60.9% on Tiny-ImageNet datasets, comparable to the baseline transformer. LL-ViT eliminates over 60% of the model weights and 50% of the multiplications in the model, and achieves 1.9x energy efficiency and 1.3x lower latency over an integer quantized ViT accelerator, while also offering superior throughput against prior works at a 10.9W power budget.

Predicting public transit incident duration from unstructured text alerts is a critical but challenging task. Addressing the domain sparsity of transit operations with standard Supervised Fine-Tuning (SFT) is difficult, as the task involves noisy, continuous labels and lacks reliable expert demonstrations for reasoning. While Reinforcement Learning from Verifiable Rewards (RLVR) excels at tasks with binary correctness, like mathematics, its applicability to noisy, continuous forecasting is an open question. This work, to our knowledge, is the first to bridge the gap between RLVR LLM training with the critical, real-world forecasting challenges in public transit operations. We adapt RLVR to this task by introducing a tolerance-based, shaped reward function that grants partial credit within a continuous error margin, rather than demanding a single correct answer. We systematically evaluate this framework on a curated dataset of NYC MTA service alerts. Our findings show that general-purpose, instruction-tuned LLMs significantly outperform specialized math-reasoning models, which struggle with the ambiguous, real-world text. We empirically demonstrate that the binary reward is unstable and degrades performance, whereas our shaped reward design is critical and allows our model to dominate on the most challenging metrics. While classical regressors are superior at minimizing overall MAE or MSE, our RLVR approach achieved a 35\% relative improvement in 5-minute accuracy (Acc@5) over the strongest baseline. This demonstrates that RLVR can be successfully adapted to real-world, noisy forecasting, but requires a verifier design that reflects the continuous nature of the problem.

Cyber-physical systems (CPS) require the joint optimization of discrete cyber actions and continuous physical parameters under stringent safety logic constraints. However, existing hierarchical approaches often compromise global optimality, whereas reinforcement learning (RL) in hybrid action spaces often relies on brittle reward penalties, masking, or shielding and struggles to guarantee constraint satisfaction. We present logic-informed reinforcement learning (LIRL), which equips standard policy-gradient algorithms with projection that maps a low-dimensional latent action onto the admissible hybrid manifold defined on-the-fly by first-order logic. This guarantees feasibility of every exploratory step without penalty tuning. Experimental evaluations have been conducted across multiple scenarios, including industrial manufacturing, electric vehicle charging stations, and traffic signal control, in all of which the proposed method outperforms existing hierarchical optimization approaches. Taking a robotic reducer assembly system in industrial manufacturing as an example, LIRL achieves a 36.47\% to 44.33\% reduction at most in the combined makespan-energy objective compared to conventional industrial hierarchical scheduling methods. Meanwhile, it consistently maintains zero constraint violations and significantly surpasses state-of-the-art hybrid-action reinforcement learning baselines. Thanks to its declarative logic-based constraint formulation, the framework can be seamlessly transferred to other domains such as smart transportation and smart grid, thereby paving the way for safe and real-time optimization in large-scale CPS.

Abstract--Federated Learning (FL) allows multiple institutions to cooperatively train machine learning models while retaining sensitive data at the source, which has great utility in privacy-sensitive environments. However, FL systems remain vulnerable to membership-inference attacks and data heterogeneity. This paper presents FedOnco-Bench, a reproducible benchmark for privacy-aware FL using synthetic oncologic CT scans with tumor annotations. Results show a distinct trade-off between privacy and utility: FedA vg is high performance (Dice around 0.85) with more privacy leakage (attack AUC about 0.72), while DP-SGD provides a higher level of privacy (AUC around 0.25) at the cost of accuracy (Dice about 0.79). FedProx and FedBN offer balanced performance under heterogeneous data, especially with non-identical distributed client data. FedOnco-Bench serves as a standardized, open-source platform for benchmarking and developing privacy-preserving FL methods for medical image segmentation. Federated Learning (FL) [1] enables multiple clients, such as hospitals, to collaboratively train machine learning models by exchanging model parameters without sharing sensitive raw data, thereby significantly enhancing privacy. FL minimizes privacy risks inherent in traditional centralized training paradigms [1]. In oncology imaging, FL has demonstrated effectiveness; for example, Alphonse et al. reported that federated models could achieve segmentation accuracy for brain tumors comparable to centrally trained models without directly sharing MRI data [2].

Structured electronic health records (EHR) are essential for clinical prediction. While count-based learners continue to perform strongly on such data, no benchmarking has directly compared them against more recent mixture-of-agents LLM pipelines, which have been reported to outperform single LLMs in various NLP tasks. In this study, we evaluated three categories of methodologies for EHR prediction using the EHRSHOT dataset: count-based models built from ontology roll-ups with two time bins, based on LightGBM and the tabular foundation model TabPFN; a pretrained sequential transformer (CLMBR); and a mixture-of-agents pipeline that converts tabular histories to natural-language summaries followed by a text classifier. We assessed eight outcomes using the EHRSHOT dataset. Across the eight evaluation tasks, head-to-head wins were largely split between the count-based and the mixture-of-agents methods. Given their simplicity and interpretability, count-based models remain a strong candidate for structured EHR benchmarking. The source code is available at: https://github.com/cristea-lab/Structured_EHR_Benchmark.

Electronic health records (EHRs) are central to modern healthcare delivery and research; yet, many researchers lack the database expertise necessary to write complex SQL queries or generate effective visualizations, limiting efficient data use and scientific discovery. To address this barrier, we introduce CELEC, a large language model (LLM)-powered framework for automated EHR data extraction and analytics. CELEC translates natural language queries into SQL using a prompting strategy that integrates schema information, few-shot demonstrations, and chain-of-thought reasoning, which together improve accuracy and robustness. On a subset of the EHRSQL benchmark, CELEC achieves execution accuracy comparable to prior systems while maintaining low latency, cost efficiency, and strict privacy by exposing only database metadata to the LLM. CELEC also adheres to strict privacy protocols: the LLM accesses only database metadata (e.g., table and column names), while all query execution occurs securely within the institutional environment, ensuring that no patient-level data is ever transmitted to or shared with the LLM. Ablation studies confirm that each component of the SQL generation pipeline, particularly the few-shot demonstrations, plays a critical role in performance. By lowering technical barriers and enabling medical researchers to query EHR databases directly, CELEC streamlines research workflows and accelerates biomedical discovery.

Measuring similarity in urban spatial networks is key to understanding cities as complex systems. Yet most existing methods are not tailored for spatial networks and struggle to differentiate them effectively. We propose GCA-Sim, a similarity-evaluation framework based on graph cellular automata. Each submodel measures similarity by the divergence between value distributions recorded at multiple stages of an information evolution process. We find that some propagation rules magnify differences among network signals; we call this "network resonance." With an improved differentiable logic-gate network, we learn several submodels that induce network resonance. We evaluate similarity through clustering performance on fifty city-level and fifty district-level road networks. The submodels in this framework outperform existing methods, with Silhouette scores above 0.9. Using the best submodel, we further observe that planning-led street networks are less internally homogeneous than organically grown ones; morphological categories from different domains contribute with comparable importance; and degree, as a basic topological signal, becomes increasingly aligned with land value and related variables over iterations.

Time series generation focuses on modeling the underlying data distribution and resampling to produce authentic time series data. Key components, such as trend and seasonality, drive temporal fluctuations, yet many existing approaches fail to employ interpretative decomposition methods, limiting their ability to synthesize meaningful trend and seasonal patterns. To address this gap, we introduce Seasonal-Trend Diffusion (STDiffusion), a novel framework for multivariate time series generation that integrates diffusion probabilistic models with advanced learnable series decomposition techniques, enhancing the interpretability of the generation process. Our approach separates the trend and seasonal learning into distinct blocks: a Multi-Layer Perceptron (MLP) structure captures the trend, while adaptive wavelet distillation facilitates effective multi-resolution learning of seasonal components. This decomposition improves the interpretability of the model on multiple scales. In addition, we designed a comprehensive correction mechanism aimed at ensuring that the generated components exhibit a high degree of internal consistency and preserve meaningful interrelationships with one another. Our empirical studies on eight real-world datasets demonstrate that STDiffusion achieves state-of-the-art performance in time series generation tasks. Furthermore, we extend the model's application to multi-window long-sequence time series generation, which delivered reliable results and highlighted its robustness and versatility.