Subscribe to the gold package and get unlimited access to Shamra Academy
Register a new userMSD: Saliency-aware Knowledge Distillation for Multimodal Understanding
MSD: تقطير المعرفة الإدراجية
288
0 0
0.0
(
0
)
Created by
Shamra Editor
Ask ChatGPT about the research
للحد من حجم النموذج ولكن الاحتفاظ بالأداء، كنا نعتمد في كثير من الأحيان على تقطير المعرفة (دينار كويتي) الذي ينقل المعرفة من نموذج المعلم الكبير إلى نموذج طالب أصغر. ومع ذلك، فإن KD على مجموعات بيانات متعددة الوسائط مثل مهام اللغة الرؤية غير مستكشفة نسبيا، وهضم معلومات متعددة الوسائط تحديا لأن طرائق مختلفة تقدم أنواعا مختلفة من المعلومات. في هذه الورقة، نقوم بإجراء دراسة تجريبية واسعة النطاق للتحقيق في أهمية وآثار كل طريقة في تقطير المعرفة. علاوة على ذلك، نقدم إطارا لتقطير المعرفة متعددة الوسائط، وقطاع التقطير الخاص بالطرياء (MSD)، لنقل المعرفة من المعلم عن مهام متعددة الوسائط عن طريق تعلم سلوك المعلم داخل كل طريقة. تهدف الفكرة إلى تحية التنبؤات الخاصة بنوية المعلم من خلال إدخال شروط الخسائر المساعدة لكل طريقة. علاوة على ذلك، نظرا لأن كل طريقة لها اتفاقية مختلفة بالنسبة للتنبؤات، فإننا نحدد درجات الرافية لكل طريقة وتحقيق في مخططات الترجيح القائم على الرافية للخسائر الإضافية. ندرس نهج تعليم الوزن لمعرفة الأثقال المثلى على شروط الخسارة هذه. في تحليلنا التجريبي، نقوم بفحص اتفاقية كل طريقة في KD، وأوضح فعالية نظام الترجيح في MSD، وإظهار أنه يحقق أداء أفضل من KD على أربعة مجموعات بيانات متعددة الوسائط.
To reduce a model size but retain performance, we often rely on knowledge distillation (KD) which transfers knowledge from a large teacher'' model to a smaller student'' model. However, KD on multimodal datasets such as vision-language tasks is relatively unexplored, and digesting multimodal information is challenging since different modalities present different types of information. In this paper, we perform a large-scale empirical study to investigate the importance and effects of each modality in knowledge distillation. Furthermore, we introduce a multimodal knowledge distillation framework, modality-specific distillation (MSD), to transfer knowledge from a teacher on multimodal tasks by learning the teacher's behavior within each modality. The idea aims at mimicking a teacher's modality-specific predictions by introducing auxiliary loss terms for each modality. Furthermore, because each modality has different saliency for predictions, we define saliency scores for each modality and investigate saliency-based weighting schemes for the auxiliary losses. We further study a weight learning approach to learn the optimal weights on these loss terms. In our empirical analysis, we examine the saliency of each modality in KD, demonstrate the effectiveness of the weighting scheme in MSD, and show that it achieves better performance than KD on four multimodal datasets.
References used
https://aclanthology.org/
rate research
Read More
Knowledge Distillation (KD) is extensively used to compress and deploy large pre-trained language models on edge devices for real-world applications. However, one neglected area of research is the impact of noisy (corrupted) labels on KD. We present,
to the best of our knowledge, the first study on KD with noisy labels in Natural Language Understanding (NLU). We document the scope of the problem and present two methods to mitigate the impact of label noise. Experiments on the GLUE benchmark show that our methods are effective even under high noise levels. Nevertheless, our results indicate that more research is necessary to cope with label noise under the KD.
In this paper we apply self-knowledge distillation to text summarization which we argue can alleviate problems with maximum-likelihood training on single reference and noisy datasets. Instead of relying on one-hot annotation labels, our student summa
rization model is trained with guidance from a teacher which generates smoothed labels to help regularize training. Furthermore, to better model uncertainty during training, we introduce multiple noise signals for both teacher and student models. We demonstrate experimentally on three benchmarks that our framework boosts the performance of both pretrained and non-pretrained summarizers achieving state-of-the-art results.
Pretrained transformer-based encoders such as BERT have been demonstrated to achieve state-of-the-art performance on numerous NLP tasks. Despite their success, BERT style encoders are large in size and have high latency during inference (especially o
n CPU machines) which make them unappealing for many online applications. Recently introduced compression and distillation methods have provided effective ways to alleviate this shortcoming. However, the focus of these works has been mainly on monolingual encoders. Motivated by recent successes in zero-shot cross-lingual transfer learning using multilingual pretrained encoders such as mBERT, we evaluate the effectiveness of Knowledge Distillation (KD) both during pretraining stage and during fine-tuning stage on multilingual BERT models. We demonstrate that in contradiction to the previous observation in the case of monolingual distillation, in multilingual settings, distillation during pretraining is more effective than distillation during fine-tuning for zero-shot transfer learning. Moreover, we observe that distillation during fine-tuning may hurt zero-shot cross-lingual performance. Finally, we demonstrate that distilling a larger model (BERT Large) results in the strongest distilled model that performs best both on the source language as well as target languages in zero-shot settings.
Recent studies argue that knowledge distillation is promising for speech translation (ST) using end-to-end models. In this work, we investigate the effect of knowledge distillation with a cascade ST using automatic speech recognition (ASR) and machin
e translation (MT) models. We distill knowledge from a teacher model based on human transcripts to a student model based on erroneous transcriptions. Our experimental results demonstrated that knowledge distillation is beneficial for a cascade ST. Further investigation that combined knowledge distillation and fine-tuning revealed that the combination consistently improved two language pairs: English-Italian and Spanish-English.
Although pre-trained big models (e.g., BERT, ERNIE, XLNet, GPT3 etc.) have delivered top performance in Seq2seq modeling, their deployments in real-world applications are often hindered by the excessive computations and memory demand involved. For ma
ny applications, including named entity recognition (NER), matching the state-of-the-art result under budget has attracted considerable attention. Drawing power from the recent advance in knowledge distillation (KD), this work presents a novel distillation scheme to efficiently transfer the knowledge learned from big models to their more affordable counterpart. Our solution highlights the construction of surrogate labels through the k-best Viterbi algorithm to distill knowledge from the teacher model. To maximally assimilate knowledge into the student model, we propose a multi-grained distillation scheme, which integrates cross entropy involved in conditional random field (CRF) and fuzzy learning.To validate the effectiveness of our proposal, we conducted a comprehensive evaluation on five NER benchmarks, reporting cross-the-board performance gains relative to competing prior-arts. We further discuss ablation results to dissect our gains.
suggested questions
Log in to be able to interact and post comments
comments
Fetching comments
Sign in to be able to follow your search criteria
