Related papers: Output Embedding Centering for Stable LLM Pretrain…

Deep convolutional neural networks are known to be unstable during training at high learning rate unless normalization techniques are employed. Normalizing weights or activations allows the use of higher learning rates, resulting in faster…

Arm end-effector stabilization is essential for humanoid loco-manipulation tasks, yet it remains challenging due to the high degrees of freedom and inherent dynamic instability of bipedal robot structures. Previous model-based controllers…

Large language Models (LLMs) have achieved significant breakthroughs across diverse domains; however, they can still produce unreliable or misleading outputs. For responsible LLM application, Uncertainty Quantification (UQ) techniques are…

Object-centric representation learning has recently been successfully applied to real-world datasets. This success can be attributed to pretrained non-object-centric foundation models, whose features serve as reconstruction targets for slot…

This work presents our ongoing research of unsupervised pretraining in neural machine translation (NMT). In our method, we initialize the weights of the encoder and decoder with two language models that are trained with monolingual data and…

Delivering meaningful uncertainty estimates is essential for a successful deployment of machine learning models in the clinical practice. A central aspect of uncertainty quantification is the ability of a model to return predictions that…

In this study, we tackle the challenge of outlier-robust predictive modeling using highly expressive neural networks. Our approach integrates two key components: (1) a transformed trimmed loss (TTL), a computationally efficient variant of…

Asynchronous reinforcement learning has become increasingly central to scaling LLM post-training, delivering major throughput gains by decoupling rollout generation from policy updates. However, widely used policy-gradient objectives such…

Adversarial training is widely used to improve the robustness of deep neural networks to adversarial attack. However, adversarial training is prone to overfitting, and the cause is far from clear. This work sheds light on the mechanisms…

Embed-to-control (E2C) is a model for solving high-dimensional optimal control problems by combining variational auto-encoders with locally-optimal controllers. However, the E2C model suffers from two major drawbacks: 1) its objective…

Reasoning failures in large language models (LLMs) are typically measured only at the end of a generation, yet many failures manifest as a process-level breakdown: the model "loses the thread" mid-reasoning. We study whether such breakdowns…

This paper proposes a composite learning backstepping control (CLBC) strategy based on modular backstepping and high-order tuners to achieve closed-loop exponential stability without high-gain feedback and PE. A novel composite learning…

Despite the significant progress made in practical applications of aligned language models (LMs), they tend to be overconfident in output answers compared to the corresponding pre-trained LMs. In this work, we systematically evaluate the…

Fine-tuning large language models (LLMs) can lead to unintended out-of-distribution generalization. Standard approaches to this problem rely on modifying training data, for example by adding data that better specify the intended…

In this work, we study the evolution of the loss Hessian across many classification tasks in order to understand the effect the curvature of the loss has on the training dynamics. Whereas prior work has focused on how different learning…

Contrastive learning has emerged as a powerful method in deep learning, excelling at learning effective representations through contrasting samples from different distributions. However, neural collapse, where embeddings converge into a…

Many safety and alignment failures of large language models (LLMs) occur due to out-of-distribution (OOD) situations: unusual prompt or response patterns that are unforeseen by model developers. We systematically study whether LLM…

Instruction fine-tuning has emerged as a critical technique for customizing Large Language Models (LLMs) to specific applications. However, recent studies have highlighted significant security vulnerabilities in fine-tuned LLMs. Existing…

Large language models are susceptible to jailbreak attacks, which can result in the generation of harmful content. While prior defenses mitigate these risks by perturbing or inspecting inputs, they ignore competing objectives, the…

Post-training alignment has increasingly become a crucial factor in enhancing the usability of language models (LMs). However, the strength of alignment varies depending on individual preferences. This paper proposes a method to incorporate…