Generative models are increasingly used to produce privacy-preserving synthetic data as a safe alternative to sharing sensitive training datasets. However, we demonstrate that such synthetic releases can still leak information about the underlying training samples through structural overlap in the data manifold. We propose a black-box membership inference attack that exploits this vulnerability without requiring access to model internals or real data. The attacker repeatedly queries the generative model to obtain large numbers of synthetic samples, performs unsupervised clustering to identify dense regions of the synthetic distribution, and then analyzes cluster medoids and neighborhoods that correspond to high-density regions in the original training data. These neighborhoods act as proxies for training samples, enabling the adversary to infer membership or reconstruct approximate records. Our experiments across healthcare, finance, and other sensitive domains show that cluster overlap between real and synthetic data leads to measurable membership leakage-even when the generator is trained with differential privacy or other noise mechanisms. The results highlight an under-explored attack surface in synthetic data generation pipelines and call for stronger privacy guarantees that account for distributional neighborhood inference rather than sample-level memorization alone, underscoring its role in privacy-preserving data publishing. Implementation and evaluation code are publicly available at:github.com/Cluster-Medoid-Leakage-Attack.

Cybersecurity's evolving complexity presents unique challenges and opportunities for AI research and practice. This paper shares key lessons and insights from designing, building, and operating production-grade GenAI pipelines in cyberse-curity, with a focus on the continual adaptation required to keep pace with ever-shifting knowledge bases, tooling, and threats. Our goal is to provide an actionable perspective for AI practitioners and industry stakeholders navigating the frontier of GenAI for cybersecurity, with particular attention to how different adaptation mechanisms complement each other in end-to-end systems. We present practical guidance derived from real-world deployments, propose best practices for leveraging retrieval-and model-level adaptation, and highlight open research directions for making GenAI more robust, precise, and auditable in cyber defense. Disclaimer: The ideas and analysis presented here are subjective. We share them based on our experience of establishing robust and efficient pipelines of generative AI for cybersecurity. In light of the age of generative AI, the objective of this document is not to provide generic descriptions of GenAI techniques, but rather to explain their practical relevance for specific contexts, and to illustrate where particular choices have worked well or poorly in our own deployments.

Machine learning (ML) models for code clone detection determine whether two pieces of code are semantically equivalent, which in turn is a key building block for software-engineering tasks like refactoring and security tasks like vulnerability and malware detection. While these models are predominantly trained on clean, structured code datasets, real-world code often undergoes a variety of semantic-preserving transformations, including refactoring, minification, automated formatting, and compiler optimizations. To address this critical gap between training and test data, we propose Auto-SPT, a novel framework to automatically construct synthetic-data generators for code. Auto-SPT is designed to produce Semantic Preserving Transformations (SPTs) that alter a program's syntactic structure while preserving its functionality and is instantiated on top of Large Language Models (LLMs). In particular, we use LLMs to craft a diverse set of SPTs, generate strong implementations for these SPTs, and compose them to result into strong transformations. Our formal analysis shows that the diversity of SPTs impacts the strength of their composition. We then empirically demonstrate that Auto-SPT generates more diverse SPTs than existing approaches and these SPTs significantly drop the performance of state-of-the-art code clone detectors. Further experiments show Auto-SPT can be used to enhance code datasets for training, to produce code-clone detection models that are robust to real-world, adversarial code transformations.

Self-driving laboratories (SDLs) have accelerated the throughput and automation capabilities for discovering and improving chemistries and materials. Although these SDLs have automated many of the steps required to conduct chemical and materials experiments, a commonly overlooked step in the automation pipeline is the handling and reloading of substrates used to transfer or deposit materials onto for downstream characterization. Here, we develop a closed-loop method of Automated Substrate Handling and Exchange (ASHE) using robotics, dual-actuated dispensers, and deep learning-driven computer vision to detect and correct errors in the manipulation of fragile and transparent substrates for SDLs. Using ASHE, we demonstrate a 98.5% first-time placement accuracy across 130 independent trials of reloading transparent glass substrates into an SDL, where only two substrate misplacements occurred and were successfully detected as errors and automatically corrected. Through the development of more accurate and reliable methods for handling various types of substrates, we move toward an improvement in the automation capabilities of self-driving laboratories, furthering the acceleration of novel chemical and materials discoveries.

This paper investigates the fundamental discontinuity between the latest two Segment Anything Models: SAM2 and SAM3. We explain why the expertise in prompt-based segmentation of SAM2 does not transfer to the multimodal concept-driven paradigm of SAM3. SAM2 operates through spatial prompts points, boxes, and masks yielding purely geometric and temporal segmentation. In contrast, SAM3 introduces a unified vision-language architecture capable of open-vocabulary reasoning, semantic grounding, contrastive alignment, and exemplar-based concept understanding. We structure this analysis through five core components: (1) a Conceptual Break Between Prompt-Based and Concept-Based Segmentation, contrasting spatial prompt semantics of SAM2 with multimodal fusion and text-conditioned mask generation of SAM3; (2) Architectural Divergence, detailing pure vision-temporal design of SAM2 versus integration of vision-language encoders, geometry and exemplar encoders, fusion modules, DETR-style decoders, object queries, and ambiguity-handling via Mixture-of-Experts in SAM3; (3) Dataset and Annotation Differences, contrasting SA-V video masks with multimodal concept-annotated corpora of SAM3; (4) Training and Hyperparameter Distinctions, showing why SAM2 optimization knowledge does not apply to SAM3; and (5) Evaluation, Metrics, and Failure Modes, outlining the transition from geometric IoU metrics to semantic, open-vocabulary evaluation. Together, these analyses establish SAM3 as a new class of segmentation foundation model and chart future directions for the emerging concept-driven segmentation era.

Assessment of medication history-taking has traditionally relied on human observation, limiting scalability and detailed performance data. While Generative AI (GenAI) platforms enable extensive data collection and learning analytics provide powerful methods for analyzing educational traces, these approaches remain largely underexplored in pharmacy clinical training. This study addresses this gap by applying learning analytics to understand how students develop clinical communication competencies with GenAI-powered virtual patients -- a crucial endeavor given the diversity of student cohorts, varying language backgrounds, and the limited opportunities for individualized feedback in traditional training settings. We analyzed 323 students' interaction logs across Australian and Malaysian institutions, comprising 50,871 coded utterances from 1,487 student-GenAI dialogues. Combining Epistemic Network Analysis to model inquiry co-occurrences with Sequential Pattern Mining to capture temporal sequences, we found that high performers demonstrated strategic deployment of information recognition behaviors. Specifically, high performers centered inquiry on recognizing clinically relevant information, integrating rapport-building and structural organization, while low performers remained in routine question-verification loops. Demographic factors including first-language background, prior pharmacy work experience, and institutional context, also shaped distinct inquiry patterns. These findings reveal inquiry patterns that may indicate clinical reasoning development in GenAI-assisted contexts, providing methodological insights for health professions education assessment and informing adaptive GenAI system design that supports diverse learning pathways.

Pose estimation of a robot arm from visual inputs is a challenging task. However, with the increasing adoption of robot arms for both industrial and residential use cases, reliable joint angle estimation can offer improved safety and performance guarantees, and also be used as a verifier to further train robot policies. This paper introduces using frontier vision-language models (VLMs) as an ``off-the-shelf" tool to estimate a robot arm's joint angles from a single target image. By evaluating frontier VLMs on both synthetic and real-world image-data pairs, this paper establishes a performance baseline attained by current FLMs. In addition, this paper presents empirical results suggesting that test time scaling or parameter scaling alone does not lead to improved joint angle predictions.

Adapting production-level computer vision tools to bespoke scientific datasets is a critical "last mile" bottleneck. Current solutions are impractical: fine-tuning requires large annotated datasets scientists often lack, while manual code adaptation costs scientists weeks to months of effort. W e consider using AI agents to automate this manual coding, and focus on the open question of optimal agent design for this targeted task. W e introduce a systematic evaluation framework for agentic code optimization and use it to study three production-level biomedical imaging pipelines. W e demonstrate that a simple agent framework consistently generates adaptation code that outperforms human-expert solutions. Our analysis reveals that common, complex agent architectures are not universally beneficial, leading to a practical roadmap for agent design.

Large language models are increasingly used to evaluate other models, yet these judgments typically lack any representation of confidence. This pilot study tests whether framing an evaluation task as a betting game (a fictional prediction market with its own LLM currency) improves forecasting accuracy and surfaces calibrated confidence signals. We generated 100 math and logic questions with verifiable answers. Six Baseline models (three current-generation, three prior-generation) answered all items. Three Predictor models then forecasted, for each question-baseline pair, if the baseline would answer correctly. Each predictor completed matched runs in two conditions: Control (simple correct/incorrect predictions) and Incentive (predictions plus wagers of 1-100,000 LLMCoin under even odds, starting from a 1,000,000 LLMCoin bankroll). Across 5,400 predictions per condition, Incentive runs showed modestly higher accuracy (81.5% vs. 79.1%, p = .089, d = 0.86) and significantly faster learning across rounds (12.0 vs. 2.9 percentage-point improvement from Round 1 to Round 4, p = .011). Most notably, stake size tracked confidence. "Whale" bets of 40,000+ coins were correct ~99% of the time, while small bets (<1,000 coins) showed only ~74% accuracy. The key finding is not that fictional money makes models smarter; accuracy gains were modest and did not reach statistical significance (p = .089) in this pilot. Rather, the betting mechanic created a legible confidence signal absent from binary yes/no outputs. This suggests that simple financial framing may help transform LLMs into risk-aware forecasters, making their internal beliefs visible and usable. The protocol offers a foundation for future work for meta-evaluation systems and what may become LLM-to-LLM prediction markets.

Analyzing underwater fish imagery is critical for ecological monitoring but remains difficult due to visual degradation and costly annotations. We introduce FishDetector-R1, a unified MLLM-based framework for fish detection, segmentation, and counting under weak supervision. On the DeepFish dataset, our framework achieves substantial gains over baselines, improving AP by 20% and mIoU by 10%, while reducing MAE by 30% and GAME by 35%. These improvements stem from two key components: a novel detect-to-count prompt that enforces spatially consistent detections and counts, and Reinforcement Learning from Verifiable Reward (RLVR) with a complementary scalable paradigm leveraging sparse point labels. Ablation studies further validate the effectiveness of this reward design. Moreover, the improvement generalizes well to other underwater datasets, confirming strong cross-domain robustness. Overall, FishDetector-R1 provides a reliable and scalable solution for accurate marine visual understanding via weak supervision. The project page for FishDetector-R1 is https://umfieldrobotics.github.io/FishDetector-R1.