Related papers: Extending MONA in Camera Dropbox: Reproduction, Le…

Reward hacking -- where RL agents exploit gaps in misspecified reward functions -- has been widely observed, but not yet systematically studied. To understand how reward hacking arises, we construct four RL environments with misspecified…

Being able to accommodate multiple simultaneous transmissions on a single channel, non-orthogonal multiple access (NOMA) appears as an attractive solution to support massive machine type communication (mMTC) that faces a massive number of…

Reasoning abilities, especially those for solving complex math problems, are crucial components of general intelligence. Recent advances by proprietary companies, such as o-series models of OpenAI, have made remarkable progress on reasoning…

Recent personalization methods for diffusion models, such as Dreambooth and LoRA, allow fine-tuning pre-trained models to generate new concepts. However, applying these techniques across consecutive tasks in order to include, e.g., new…

Various communication technologies are expected to utilize mobile ad hoc networks (MANETs). By combining MANETs with non-orthogonal multiple access (NOMA) communications, one can support scalable, spectrally efficient, and flexible network…

Multimodal models are ubiquitous, yet existing explainability methods are often single-modal, architecture-dependent, or too computationally expensive to run at scale. We introduce LEMON (Local Explanations via Modality-aware OptimizatioN),…

Designing a good reward function is essential to robot planning and reinforcement learning, but it can also be challenging and frustrating. The reward needs to work across multiple different environments, and that often requires many…

Non-orthogonal multiple access (NOMA) is a powerful transmission technique that enhances the spectral efficiency of communication links, and is being investigated for 5G standards and beyond. A major drawback of NOMA is the need to apply…

The rising popularity of large foundation models has led to a heightened demand for parameter-efficient fine-tuning methods, such as Low-Rank Adaptation (LoRA), which offer performance comparable to full model fine-tuning while requiring…

Embodied AI agents exploit reward signal flaws through reward hacking, achieving high proxy scores while failing true objectives. We introduce Mechanistically Interpretable Task Decomposition (MITD), a hierarchical transformer architecture…

Using parallel embedded systems these days is increasing. They are getting more complex due to integrating multiple functionalities in one application or running numerous ones concurrently. This concerns a wide range of applications,…

Current SOTA adversarially robust models are mostly based on adversarial training (AT) and differ only by some regularizers either at inner maximization or outer minimization steps. Being repetitive in nature during the inner maximization…

In this work, we study optimization methods that leverage the linear minimization oracle (LMO) over a norm-ball. We propose a new stochastic family of algorithms that uses the LMO to adapt to the geometry of the problem and, perhaps…

Current state-of-the-art Neural Architecture Search (NAS) methods neither efficiently scale to multiple hardware platforms, nor handle diverse architectural search-spaces. To remedy this, we present DONNA (Distilling Optimal Neural Network…

Big data repositories from online learning platforms such as Massive Open Online Courses (MOOCs) represent an unprecedented opportunity to advance research on education at scale and impact a global population of learners. To date, such…

Preference alignment is pivotal for empowering large language models (LLMs) to generate helpful and harmless responses. However, the performance of preference alignment is highly sensitive to the prevalent noise in the preference data.…

Recent works on domain adaptation exploit adversarial training to obtain domain-invariant feature representations from the joint learning of feature extractor and domain discriminator networks. However, domain adversarial methods render…

Recent studies have demonstrated significant progress in aligning text-to-image diffusion models with human preference via Reinforcement Learning from Human Feedback. However, while existing methods achieve high scores on automated reward…

Reinforcement learning (RL) commonly relies on scalar rewards with limited ability to express temporal, conditional, or safety-critical goals, and can lead to reward hacking. Temporal logic expressible via the more general class of…

Cross-Domain Recommendation (CDR) stands as a pivotal technology addressing issues of data sparsity and cold start by transferring general knowledge from the source to the target domain. However, existing CDR models suffer limitations in…