Exploring the Limits of Simple Learners in Knowledge Distillation for Document Classification with DocBERT
Fine-tuned variants of BERT are able to achieve state-of-the-art accuracy on many natural language processing tasks, although at significant computational costs. In this paper, we verify BERT{'}s effectiveness for document classification and investigate the extent to which BERT-level effectiveness can be obtained by different baselines, combined with knowledge distillation{---}a popular model compression method. The results show that BERT-level effectiveness can be achieved by a single-layer LSTM with at least $40\times$ fewer FLOPS and only ${\sim}3\%$ parameters. More importantly, this study analyzes the limits of knowledge distillation as we distill BERT{'}s knowledge all the way down to linear models{---}a relevant baseline for the task. We report substantial improvement in effectiveness for even the simplest models, as they capture the knowledge learnt by BERT.
Code (0)
등록된 구현이 없습니다.
Tasks
Document ClassificationGeneral ClassificationKnowledge DistillationModel CompressionMethods 이 논문이 사용한 방법론
Similar Papers 제목 키워드 기반
Linkless Link Prediction via Relational Distillation
Graph Neural Networks (GNNs) have shown exceptional performance in the task of link prediction. Despite their effectiveness, the high latency brought by non-trivial neighborhood data dependency limits GNNs in practical d…
Knowledge DistillationLink PredictionNode ClassificationPredictionLearning Behavior-oriented Knowledge Tracing
Exploring how learners' knowledge states evolve during the learning activities is a critical task in online learning systems, which can facilitate personalized services downstream, such as course recommendation. Most of …
Knowledge TracingCooperative Knowledge Distillation: A Learner Agnostic Approach
Knowledge distillation is a simple but powerful way to transfer knowledge between a teacher model to a student model. Existing work suffers from at least one of the following key limitations in terms of direction and sco…
counterfactualKnowledge DistillationSelf-Supervised LearningTransfer LearningLeveraging Graph Retrieval-Augmented Generation to Support Learners' Understanding of Knowledge Concepts in MOOCs
Massive Open Online Courses (MOOCs) lack direct interaction between learners and instructors, making it challenging for learners to understand new knowledge concepts. Recently, learners have increasingly used Large Langu…
Knowledge GraphsQuestion AnsweringQuestion GenerationQuestion-Generation+2Exploring the Limits of Model Compression in LLMs: A Knowledge Distillation Study on QA Tasks
Large Language Models (LLMs) have demonstrated outstanding performance across a range of NLP tasks, however, their computational demands hinder their deployment in real-world, resource-constrained environments. This work…
Knowledge DistillationQuestion AnsweringModel Compression