Papers with DistilBERT
UnMASKed: Quantifying Gender Biases in Masked Language Models through Linguistically Informed Job Market Prompts (2024.eacl-srw)
Copied to clipboard
| Challenge: | Language models (LMs) often include societal biases encoded in the human-produced datasets used for their training. |
| Approach: | They evaluated six prominent language models: BERT, RoBERTa, DistilBERT, BERT- multilingual, XLM-RoBERT and DistilberT- multilinguistic. |
| Outcome: | The results show that the models generated by the models were stereotypically gendered and with a reduced bias in multilingual variants. |
How Are Idioms Processed Inside Transformer Language Models? (2023.starsem-1)
Copied to clipboard
| Challenge: | idioms are prevalent in natural language, but how do they be processed? |
| Approach: | They analyze the embeddings of idiomatic and literal expressions across all layers of the networks at both the sentence and word levels. |
| Outcome: | The proposed models represent idioms distinctively compared to literal language, the study finds . |
Accelerating BERT Inference for Sequence Labeling via Early-Exit (2021.acl-long)
Copied to clipboard
| Challenge: | Existing early-exit mechanisms are designed for sequence-level tasks, rather than sequence labeling. |
| Approach: | They propose to extend sentence-level early-exit to accelerate inference of PTMs . they propose a token-level mechanism that allows partial tokens to exit early at different layers . |
| Outcome: | The proposed approach can save up to 66%75% inference cost with minimal performance degradation. |
Language as a fingerprint: Self-supervised learning of user encodings using transformers (2022.findings-emnlp)
Copied to clipboard
| Challenge: | Existing studies on individual differences and language representations focused on predicting selected attributes from text or conditioning text representations on author attributes. |
| Approach: | They propose a self-supervised approach to learning language-based user encodings using transformers. |
| Outcome: | The proposed model can pick up on complex linguistic signatures of users and infer rich information about them. |
Should I try multiple optimizers when fine-tuning a pre-trained Transformer for NLP tasks? Should I tune their hyperparameters? (2024.eacl-long)
Copied to clipboard
| Challenge: | NLP research has explored different neural model architectures and sizes, datasets, training objectives, and transfer-learning techniques. |
| Approach: | They propose to use a variant of Stochastic Gradient Descent (SGD) to select among numerous variants, often with minimal or no tuning of the optimizer’s hyperparameters. |
| Outcome: | Experiments with five GLUE datasets, two models and seven popular optimizers show that tuning just the learning rate is as good as tuning all the hyperparameters. |
Masking as an Efficient Alternative to Finetuning for Pretrained Language Models (2020.emnlp-main)
Copied to clipboard
| Challenge: | Extensive evaluations of masking BERT, RoBERTa, and DistilBERT on eleven diverse NLP tasks show that our binary masked language models encode information necessary for solving downstream tasks. |
| Approach: | They propose an efficient method of utilizing pretrained language models where selective binary masks are learned instead of finetuning. |
| Outcome: | Extensive evaluations of masking BERT, RoBERTa, and DistilBERT on eleven diverse NLP tasks show that the proposed method yields comparable performance to finetuning, but has a much smaller memory footprint when multiple tasks need to be solved. |
Towards a Comprehensive Understanding and Accurate Evaluation of Societal Biases in Pre-Trained Transformers (2021.naacl-main)
Copied to clipboard
| Challenge: | Existing pre-trained language models are not fully considered for societal biases . pre-training models can be useful for many NLP tasks, but they can be harmful when used at scale. |
| Approach: | They investigate gender and racial bias across pre-trained language models . they evaluate bias within pre-trainers using three metrics: WEAT, sequence likelihood, and pronoun ranking. |
| Outcome: | The proposed model fails to detect gender and racial biases in pre-trained models . the model is ineffective when word embedding, demonstrating the need for more robust bias testing in transformers. |
At the Intersection of NLP and Sustainable Development: Exploring the Impact of Demographic-Aware Text Representations in Modeling Value on a Corpus of Interviews (2022.lrec-1)
Copied to clipboard
| Challenge: | In order to preserve the privacy of speakers, we investigate encoding demographic information using autoencoders. |
| Approach: | They introduce a dataset of qualitative interviews from rural communities in India and Uganda and use it to enhance text representations with demographic information. |
| Outcome: | The proposed model extends the UPV classification model with demographic information to preserve the privacy of speakers. |
Bag-of-Words vs. Graph vs. Sequence in Text Classification: Questioning the Necessity of Text-Graphs and the Surprising Strength of a Wide MLP (2022.acl-long)
Copied to clipboard
| Challenge: | Existing graph-based text classification methods are based on a Bag-of-Words (BoW) . |
| Approach: | They propose to use a Bag-of-Words-based multi-layer perceptron model to perform inductive text classification using a sequence-based BERT and a lightweight DistilBERT model to fine-tune their results. |
| Outcome: | The proposed model outperforms the current graph-based models and questions the importance of synthetic graphs in modern text classification methods. |
Efficient Transformer-based Large Scale Language Representations using Hardware-friendly Block Structured Pruning (2020.findings-emnlp)
Copied to clipboard
| Challenge: | Pretrained large-scale language models have been criticized for their limited weight storage and computational speed on hardware platforms. |
| Approach: | They propose an efficient transformer-based large-scale language representation using hardware-friendly block structure pruning. |
| Outcome: | The proposed model achieves 5.0x accuracy on GLUE benchmarks and 1.79x compression rate on DistilBERT. |
AutoTinyBERT: Automatic Hyper-parameter Optimization for Efficient Pre-trained Language Models (2021.acl-long)
Copied to clipboard
| Challenge: | Pre-trained language models (PLMs) have achieved great success in natural language processing. |
| Approach: | They propose a method that automatically searches architecture hyper-parameters in BERT . they use one-shot learning and the search space to provide an adaptive development way . |
| Outcome: | The proposed method outperforms both the baseline and distillation-based methods on GLUE and SQUAD benchmarks. |
Adaptive Parameter Compression for Language Models (2025.findings-naacl)
Copied to clipboard
| Challenge: | Adaptive parameter compression is a new approach to improve NLP models . the current algorithm is based on a single parameter, but it is not scalable. |
| Approach: | They propose a hardware-independent compression strategy that extends the weight-squeezing approach by introducing compression biases and weights. |
| Outcome: | The proposed compression strategy outperforms DistilBERT base models while being significantly more efficient. |
Compressing Transformer-Based Semantic Parsing Models using Compositional Code Embeddings (2020.findings-emnlp)
Copied to clipboard
Prafull Prakash, Saurabh Kumar Shashidhar, Wenlong Zhao, Subendhu Rongali, Haidar Khan, Michael Kayser
| Challenge: | Existing task-oriented semantic parsing models use BERT or RoBERTa as pretrained encoders. |
| Approach: | They propose to learn compositional code embeddings to greatly reduce the sizes of BERT and RoBERTa encoders. |
| Outcome: | The proposed model reduces the size of BERT and RoBERTa encoders while maintaining performance. |
ALBETO and DistilBETO: Lightweight Spanish Language Models (2022.lrec-1)
Copied to clipboard
| Challenge: | Recent advances in pre-trained language models have made them more popular . however, there are still limited versions of these models for other languages . |
| Approach: | They present ALBETO and DistilBETO which are versions of ALBERT and DistillBERT pre-trained exclusively on Spanish corpora. |
| Outcome: | The proposed models outperform BETO and ALBERT on Spanish datasets . the models outpace BETO on MLDoc, PAWS-X, XNLI, MLQA, SQAC and XQuAD datasets. |
Neural Mask Generator: Learning to Generate Adaptive Word Maskings for Language Model Adaptation (2020.emnlp-main)
Copied to clipboard
| Challenge: | Existing methods to train language models on diverse text corpora have brought up performance improvements on several natural language understanding (NLU) tasks. |
| Approach: | They propose a method to automatically generate domain- and task-adaptive maskings of a given text for self-supervised pre-training. |
| Outcome: | The proposed framework outperforms rule-based masking strategies on question answering and text classification datasets on which it outperformed rule-driven masking techniques. |
Compositional and Lexical Semantics in RoBERTa, BERT and DistilBERT: A Case Study on CoQA (2020.emnlp-main)
Copied to clipboard
| Challenge: | Existing knowledge transfer models do not exploit compositionality of language, often relying on superficial features. |
| Approach: | They propose to use a knowledge distillation technique to fine tune RoBERTa, BERT and DistilBERT models to improve their performance. |
| Outcome: | The proposed models improve on the CoQA task with linguistic knowledge and are able to represent compositional and lexical information. |
Attention-Enhancing Backdoor Attacks Against BERT-based Models (2023.findings-emnlp)
Copied to clipboard
| Challenge: | Existing textual backdoor attacks focus on generating stealthy triggers or modifying model weights. |
| Approach: | They propose a Trojan Attention Loss (TAL) which enhances the Trojan behavior by directly manipulating attention patterns. |
| Outcome: | The proposed method improves the effectiveness of the backdoor attacks on different backbone models and tasks. |
VE-KD: Vocabulary-Expansion Knowledge-Distillation for Training Smaller Domain-Specific Language Models (2024.findings-emnlp)
Copied to clipboard
| Challenge: | VE-KD is a method that balances knowledge distillation and vocabulary expansion with the aim of training efficient domain-specific language models. |
| Approach: | They propose a method that balances knowledge distillation and vocabulary expansion with the aim of training efficient domain-specific language models. |
| Outcome: | VE-KD outperforms DistilBERT and Adapt-and-Distill in biomedical domain tasks . compared with other methods, it outperformed Distilbert and adapted-and distill . |