相关论文: Rethinking the Comparison Unit in Sequence-Level R…

Self-supervised learning (SSL) has proven vital in speech and audio-related applications. The paradigm trains a general model on unlabeled data that can later be used to solve specific downstream tasks. This type of model is costly to train…

We address an important problem in sequence-to-sequence (Seq2Seq) learning referred to as copying, in which certain segments in the input sequence are selectively replicated in the output sequence. A similar phenomenon is observable in…

Understanding whether fine-tuning elicits latent capabilities or teaches new ones is a fundamental question for language model evaluation and safety. We develop a formal information-theoretic framework for quantifying how much predictive…

We propose a novel contrastive learning framework to effectively address the challenges of data heterogeneity in federated learning. We first analyze the inconsistency of gradient updates across clients during local training and establish…

Reinforcement learning (RL) has been pivotal in enhancing the reasoning capabilities of large language models (LLMs), but it often suffers from limited exploration and entropy collapse, where models exploit a narrow set of solutions,…

Comparative reasoning is a process of comparing objects, concepts, or entities to draw conclusions, which constitutes a fundamental cognitive ability. In this paper, we propose a novel framework to pre-train language models for enhancing…

Reinforcement learning (RL) has emerged as a powerful post-training paradigm for enhancing the reasoning capabilities of large language models (LLMs). However, reinforcement learning for LLMs faces substantial data scarcity challenges,…

Transformer-based large language models (LLMs) have achieved remarkable success, yet their standard attention mechanism incurs quadratic computation and memory costs with respect to sequence length, posing a major bottleneck for…

The increasing complexity of modern deep neural network models and the expanding sizes of datasets necessitate the development of optimized and scalable training methods. In this white paper, we addressed the challenge of efficiently…

We consider the problem of length generalization in sequence prediction. We define a new metric of performance in this setting -- the Asymmetric-Regret -- which measures regret against a benchmark predictor with longer context length than…

In this work, we study the issue of reward hacking on the response length, a challenge emerging in Reinforcement Learning from Human Feedback (RLHF) on LLMs. A well-formatted, verbose but less helpful response from the LLMs can often…

In this work, we aim to establish a strong connection between two significant bodies of machine learning research: continual learning and sequence modeling. That is, we propose to formulate continual learning as a sequence modeling problem,…

Recent works in multiple object tracking use sequence model to calculate the similarity score between the detections and the previous tracklets. However, the forced exposure to ground-truth in the training stage leads to the…

Lifelong learning remains an open problem. One of its main difficulties is catastrophic forgetting. Many dynamic expansion approaches have been proposed to address this problem, but they all use homogeneous models of predefined structure…

To better align Large Language Models (LLMs) with human judgment, Reinforcement Learning from Human Feedback (RLHF) learns a reward model and then optimizes it using regularized RL. Recently, direct alignment methods were introduced to…

Reinforcement learning with verifiable rewards (RLVR) has become a central paradigm for improving reasoning and code generation in large language models, and GRPO-style training is widely adopted for its simplicity and effectiveness.…

Consider two data providers, each maintaining records of different feature sets about common entities. They aim to learn a linear model over the whole set of features. This problem of federated learning over vertically partitioned data…

Most recent state of the art architectures rely on combinations and variations of three approaches: convolutional, recurrent and self-attentive methods. Our work attempts in laying the basis for a new research direction for sequence…

Recent advances in reasoning domains with neural networks have primarily been enabled by a training recipe that optimizes Large Language Models, previously trained to predict the next-token in a sequence, with reinforcement learning…

Sequence comparison is a widely used computational technique in modern molecular biology. In spite of the frequent use of sequence comparisons the important problem of assigning statistical significance to a given degree of similarity is…