Abstract We propose Conditional Adapter (CoDA), a parameter-efficient transfer learning
method that also improves inference efficiency. CoDA generalizes beyond standard adapter …

Speculative decoding (SD) accelerates large language model inference by employing a
faster draft model for generating multiple tokens, which are then verified in parallel by the …

Knowledge distillation is commonly used for compressing neural networks to reduce their
inference cost and memory footprint. However, current distillation methods for auto …

Large language models have become increasingly difficult to train because of the growing
computation time and cost. In this work, we present SRU++, a highly-efficient architecture …

Reinforcement learning (RL) has emerged as a powerful paradigm for fine-tuning Large
Language Models (LLMs) for conditional text generation. In particular, recent LLMs such as …

Planning appropriate driving trajectory for route following is an important function for
autonomous driving. Behavioral cloning, which allows automatic trajectory learning and …

Knowledge distillation (KD) is widely used for compressing a teacher model to a smaller
student model, reducing its inference cost and memory footprint while preserving model …

Knowledge distillation (KD) is an effective strategy for neural machine translation (NMT) to
improve the performance of a student model. Usually, the teacher can guide the student to …

This paper demonstrates that by fine-tuning an autoregressive language model (GPT-Neo)
on appropriately structured step-by-step demonstrations, it is possible to teach it to execute a …

Document-level machine translation faces the challenge of data sparsity due to its long input
length and a small amount of training data, increasing the risk of learning spurious patterns …