Bitune: Bidirectional Instruction-Tuning
We introduce Bitune, a method that improves instruction-tuning of pretrained decoder-only large language models, leading to consistent gains on downstream tasks. Bitune applies both causal and bidirectional attention to the prompt, to obtain a better representation of the query or instruction. We realize this by introducing two sets of parameters, for which we apply parameter-efficient finetuning techniques. These causal and bidirectional features are then combined into a weighted average with trainable coefficients, which is subsequently used to generate new tokens. We demonstrate significant improvements in zero-shot performance on commonsense reasoning, arithmetic, and language understanding tasks, while extensive ablation studies validate the role of each component and demonstrate the method's agnosticism to different PEFT techniques.
Code (0)
등록된 구현이 없습니다.
Tasks
DecoderSimilar Papers 제목 키워드 기반
InstructProtein: Aligning Human and Protein Language via Knowledge Instruction
Large Language Models (LLMs) have revolutionized the field of natural language processing, but they fall short in comprehending biological sequences such as proteins. To address this challenge, we propose InstructProtein…
Knowledge GraphsProtein Function PredictionText GenerationImproved Large Language Diffusion Models
Modern large language models are predominantly trained with autoregressive factorization and causal attention. We present \emph{iLLaDA}, an 8B masked diffusion language model trained from scratch with fully bidirectional…
Unlock the Power: Competitive Distillation for Multi-Modal Large Language Models
Recently, multi-modal content generation has attracted lots of attention from researchers by investigating the utilization of visual instruction tuning based on large language models (LLMs). To enhance the performance an…
Knowledge DistillationTransfer LearningBuilding Open-Ended Embodied Agent via Language-Policy Bidirectional Adaptation
Building embodied agents on integrating Large Language Models (LLMs) and Reinforcement Learning (RL) have revolutionized human-AI interaction: researchers can now leverage language instructions to plan decision-making fo…
Decision MakingLanguage ModellingReinforcement Learning (RL)Supervised In-Context Fine-Tuning for Generative Sequence Labeling
Sequence labeling (SL) tasks, where labels are assigned to tokens, are abundant in NLP (e.g., named entity recognition and aspect-based sentiment analysis). Owing to the intuition that they require bidirectional context,…
Response GenerationSentiment Analysis