Having the ability to infer characteristics of autonomous agents would profoundly revolutionize defense, security, and civil applications. Our previous work was the first to demonstrate that supervised neural network time series classification (NN TSC) could rapidly predict the tactics of swarming autonomous agents in military contexts, providing intelligence to inform counter-maneuvers. However, most autonomous interactions, especially military engagements, are fraught with uncertainty, raising questions about the practicality of using a pretrained classifier. This article addresses that challenge by leveraging expected operational variations to construct a richer dataset, resulting in a more robust NN with improved inference performance in scenarios characterized by significant uncertainties. Specifically, diverse datasets are created by simulating variations in defender numbers, defender motions, and measurement noise levels. Key findings indicate that robust NNs trained on an enriched dataset exhibit enhanced classification accuracy and offer operational flexibility, such as reducing resources required and offering adherence to trajectory constraints. Furthermore, we present a new framework for optimally deploying a trained NN by the defenders. The framework involves optimizing defender trajectories that elicit adversary responses that maximize the probability of correct NN tactic classification while also satisfying operational constraints imposed on the defenders.

Distinguishing in- and out-of-distribution (OOD) inputs is crucial for reliable deployment of classification systems. However, OOD data is typically unavailable or difficult to collect, posing a significant challenge for accurate OOD detection. In this work, we present a method that harnesses the generative capabilities of Large Language Models (LLMs) to create high-quality synthetic OOD proxies, eliminating the dependency on any external OOD data source. We study the efficacy of our method on classical text classification tasks such as toxicity detection and sentiment classification as well as classification tasks arising in LLM development and deployment, such as training a reward model for RLHF and detecting misaligned generations. Extensive experiments on nine InD-OOD dataset pairs and various model sizes show that our approach dramatically lowers false positive rates (achieving a perfect zero in some cases) while maintaining high accuracy on in-distribution tasks, outperforming baseline methods by a significant margin.

Despite the generative model's groundbreaking success, the need to study its implications for privacy and utility becomes more urgent. Although many studies have demonstrated the privacy threats brought by GANs, no existing survey has systematically categorized the privacy and utility perspectives of GANs and VAEs. In this article, we comprehensively study privacy-preserving generative models, articulating the novel taxonomies for both privacy and utility metrics by analyzing 100 research publications. Finally, we discuss the current challenges and future research directions that help new researchers gain insight into the underlying concepts.

Out-of-distribution (OOD) detection is crucial for developing trustworthy and reliable machine learning systems. Recent advances in training with auxiliary OOD data demonstrate efficacy in enhancing detection capabilities. Nonetheless, these methods heavily rely on acquiring a large pool of high-quality natural outliers. Some prior methods try to alleviate this problem by synthesizing virtual outliers but suffer from either poor quality or high cost due to the monotonous sampling strategy and the heavy-parameterized generative models. In this paper, we overcome all these problems by proposing the Hamiltonian Monte Carlo Outlier Synthesis (HamOS) framework, which views the synthesis process as sampling from Markov chains. Based solely on the in-distribution data, the Markov chains can extensively traverse the feature space and generate diverse and representative outliers, hence exposing the model to miscellaneous potential OOD scenarios. The Hamiltonian Monte Carlo with sampling acceptance rate almost close to 1 also makes our framework enjoy great efficiency. By empirically competing with SOTA baselines on both standard and large-scale benchmarks, we verify the efficacy and efficiency of our proposed HamOS.

Despite numerous attempts to further improve memory efficiency (Mitzenmacher, 2018; Dai & Recent studies have demonstrated that learned Shrivastava, 2020), existing LBFs face two critical unresolved Bloom filters, which combine machine learning issues: (1) the balance between the machine learning with the classical Bloom filter, can achieve superior model size and the Bloom filter size remains suboptimal, memory efficiency. However, existing learned and (2) the reject time cannot be effectively minimized. Bloom filters face two critical unresolved challenges: the balance between the machine learning (1) The Balance Between Machine Learning Model Size model size and the Bloom filter size is not optimal, and Bloom Filter Size: Existing LBFs lack mechanisms and the reject time cannot be minimized to automatically balance the sizes of the machine learning effectively. We propose the Cascaded Learned model and the Bloom filters. An LBF consists of a machine Bloom Filter (CLBF) to address these issues. Our learning model and one or more Bloom filters, aiming to dynamic programming-based optimization automatically minimize the total memory usage, i.e., the sum of the memory selects configurations that achieve an consumed by the machine learning model and the Bloom optimal balance between the model and filter sizes filters. Since a smaller machine learning model often--but while minimizing reject time. Experiments on not always--has lower accuracy, larger Bloom filters are real-world datasets show that CLBF reduces memory needed to maintain the overall false positive rate of an LBF, usage by up to 24% and decreases reject time whereas a larger model often--but not always--allows for by up to 14 times compared to state-of-the-art smaller Bloom filters. Thus, it is a challenging task to strike learned Bloom filters.

AI models might use deceptive strategies as part of scheming or misaligned behaviour. Monitoring outputs alone is insufficient, since the AI might produce seemingly benign outputs while their internal reasoning is misaligned. We thus evaluate if linear probes can robustly detect deception by monitoring model activations. We test two probe-training datasets, one with contrasting instructions to be honest or deceptive (following Zou et al., 2023) and one of responses to simple roleplaying scenarios. We test whether these probes generalize to realistic settings where Llama-3.3-70B-Instruct behaves deceptively, such as concealing insider trading (Scheurer et al., 2023) and purposely underperforming on safety evaluations (Benton et al., 2024). We find that our probe distinguishes honest and deceptive responses with AUROCs between 0.96 and 0.999 on our evaluation datasets. If we set the decision threshold to have a 1% false positive rate on chat data not related to deception, our probe catches 95-99% of the deceptive responses. Overall we think white-box probes are promising for future monitoring systems, but current performance is insufficient as a robust defence against deception. Our probes' outputs can be viewed at data.apolloresearch.ai/dd and our code at github.com/ApolloResearch/deception-detection.

Approximately 10% of newborns need some assistance to start breathing and 5\% proper ventilation. It is crucial that interventions are initiated as soon as possible after birth. Accurate documentation of Time of Birth (ToB) is thereby essential for documenting and improving newborn resuscitation performance. However, current clinical practices rely on manual recording of ToB, typically with minute precision. In this study, we present an AI-driven, video-based system for automated ToB detection using thermal imaging, designed to preserve the privacy of healthcare providers and mothers by avoiding the use of identifiable visual data. Our approach achieves 91.4% precision and 97.4% recall in detecting ToB within thermal video clips during performance evaluation. Additionally, our system successfully identifies ToB in 96% of test cases with an absolute median deviation of 1 second compared to manual annotations. This method offers a reliable solution for improving ToB documentation and enhancing newborn resuscitation outcomes.

Residual transformation is critical to improving representational depth and expressive power of large language models (LLMs). However, the use of static residual transformations across all tokens during auto-regressive generation induces a suboptimal balance between inference efficiency and generation fidelity. Existing methods, including Early Exiting, Skip Decoding, and Mixture-of-Depth, attempt to address this by modulating the residual transformation based on token-level complexity. Nevertheless, these approaches predominantly consider the distance traversed by tokens through the model layers, neglecting the underlying velocity of residual evolution. In this work, we introduce Mixture of Multi-rate Residuals, a novel framework that dynamically modulates the velocity of residual transformations to optimize early residual alignment. This modification improves inference efficiency by better aligning intermediate representations at earlier stages. We show the efficacy of our technique in diverse optimization setups such as dynamic computing, speculative decoding, and MoE Ahead-of-Time (AoT) loading using challenging reasoning tasks from Koala, Self-Instruct, WizardLM and MT Bench. Our approach empirically outperforms state-of-the-art distance-based residual strategies, enabling a better trade-off between generation metrics and speedup in dynamic computing settings. In self-speculative decoding setups, M2R2 achieves up to 2.8X speedups on MT-Bench under lossless conditions, outperforming SOTA approaches such as 2-model speculative decoding, Medusa, LookAhead Decoding, and DEED. In Mixture-of-Experts (MoE) architectures, we enhance decoding speed by coupling early residual alignment with ahead-oftime expert loading into high-bandwidth memory (HBM). This enables concurrent memory access and computation, reducing the latency bottlenecks inherent in expert switching during decoding. Empirical results show that our method delivers a speedup of 2.9X in MoE architectures, positioning it as a highly effective strategy in resource-constrained environments.

This technology has important applications in many challenges of the industry such as content moderation, tracing AI-generated content, and monitoring the usage of AI models. The contributions of this thesis include the development of new watermarking techniques for images, audio, and text. We first introduce methods for active moderation of images on social platforms. We then develop specific techniques for AI-generated content. We specifically demonstrate methods to adapt latent generative models to embed watermarks in all generated content, identify watermarked sections in speech, and improve watermarking in large language models with tests that ensure low false positive rates. Furthermore, we explore the use of digital watermarking to detect model misuse, including the detection of watermarks in language models fine-tuned on watermarked text, and introduce training-free watermarks for the weights of large transformers. Through these contributions, the thesis provides effective solutions for the challenges posed by the increasing use of generative AI models and the need for model monitoring and content moderation. It finally examines the challenges and limitations of watermarking techniques and discuss potential future directions for research in this area.

Spiking Neural Networks (SNNs) mimic biological brain activity, processing data efficiently through an event-driven design, wherein the neurons activate only when inputs exceed specific thresholds. Their ability to track voltage changes over time via membrane potential dynamics helps retain temporal information. This study combines SNNs with PyTorch's adaptable framework, snnTorch, to test their potential for image-based tasks. We introduce DRiVE, a vehicle detection model that uses spiking neuron dynamics to classify images, achieving 94.8% accuracy and a near-perfect 0.99 AUC score. These results highlight DRiVE's ability to distinguish vehicle classes effectively, challenging the notion that SNNs are limited to temporal data. As interest grows in energy-efficient neural models, DRiVE's success emphasizes the need to refine SNN optimization for visual tasks. This work encourages broader exploration of SNNs in scenarios where conventional networks struggle, particularly for real-world applications requiring both precision and efficiency.