Are Larger Pretrained Language Models Uniformly Better? Comparing Performance at the Instance Level (2021.findings-acl)
Copied to clipboard
| Challenge: | Larger models have higher out-of-distribution robustness, while smaller models have lower accuracy on rare subgroups. |
| Approach: | They develop statistically rigorous methods to investigate whether large models are better on every instance . they find that individual predictions are highly sensitive to noise in the randomness in training . |
| Outcome: | The proposed model is worse than BERT-MINI on 1-4% of instances across MNLI, SST-2, and QQP, compared to the overall accuracy improvement of 2-10%. |
Similar Papers
Contextual Embeddings: When Are They Worth It? (2020.acl-main)
Copied to clipboard
| Challenge: | In recent years, rich contextual embeddings have enabled rapid progress on benchmarks like GLUE, but require significant computational resources during pretraining and during downstream task training and inference. |
| Approach: | They empirically compare contextual embeddings with classic pretrained embedders and a random word embeddable with a simple baseline. |
| Outcome: | The proposed models perform within 5 to 10% accuracy on industry-scale data. |
On the importance of pre-training data volume for compact language models (2020.emnlp-main)
Copied to clipboard
| Challenge: | Recent advances in language modeling have led to computationally intensive and resource-demanding state-of-the-art models. |
| Approach: | They investigate the impact of pre-training data volume on compact language models . they use a French question answering task to train models with as little as 100 MB of text . |
| Outcome: | The results show that pre-training data volume can improve models with as little as 100 MB of text . the results suggest that the model performance is poorer with less data than with larger datasets . |
Are Compressed Language Models Less Subgroup Robust? (2023.emnlp-main)
Copied to clipboard
| Challenge: | Existing methods to reduce model size and latency while retaining overall performance are not known about their impact on subgroup robustness. |
| Approach: | They investigate the effects of model compression on subgroup robustness of BERT language models. |
| Outcome: | The proposed compression methods do not worsen the performance on minority subgroups. |
A Comparison between Pre-training and Large-scale Back-translation for Neural Machine Translation (2021.findings-acl)
Copied to clipboard
| Challenge: | BERT is a promising technique to improve NMT, but how it outperforms standard NMT is understudied. |
| Approach: | We compare MT engines trained with pre-trained BERT and back-translation with incrementally larger amounts of data. |
| Outcome: | The proposed technique outperforms standard NMT models on morphology and syntax. |
Investigating Ensemble Methods for Model Robustness Improvement of Text Classifiers (2022.findings-emnlp)
Copied to clipboard
| Challenge: | Existing methods to reduce model's reliance on bias features ignore the learnability of these features. |
| Approach: | They propose to reduce models' reliance on bias features by first training models with fixed low-capacity models which ignore the learnability of the bias features. |
| Outcome: | The proposed models can perform better on out-of-distribution datasets than baseline models with a more sophisticated model design. |
Revisiting Scaling Laws for Language Models: The Role of Data Quality and Training Strategies (2025.acl-long)
Copied to clipboard
| Challenge: | Existing scaling laws suggest augmenting model size and training data results in enhanced performance, but recent studies reveal deviations, particularly in large language models, where performance improvements decelerate—a phenomenon known as sub-scaling. |
| Approach: | They propose a sub-optimal scaling law that better predicts performance in sub-scaling regimes by examining data quality and training strategies. |
| Outcome: | The proposed scaling law better predicts performance in sub-scaling regimes, highlighting the importance of data quality and diversity. |
Methods for Estimating and Improving Robustness of Language Models (2022.naacl-srw)
Copied to clipboard
| Challenge: | Large language models suffer from weak generalisation ability due to shallow textual relations over full semantic complexity of the problem. |
| Approach: | They propose to incorporate some of these measures into training objectives to enhance distributional robustness of LLMs. |
| Outcome: | The proposed models outperform human models on complex tasks and outperformed other models on deep networks. |
Deciphering the Impact of Pretraining Data on Large Language Models through Machine Unlearning (2024.findings-acl)
Copied to clipboard
| Challenge: | Existing studies have suggested that the composition of the pretraining corpus exerts a significant impact upon the performance of LLMs. |
| Approach: | They analyze the impact of 48 datasets from 5 major categories of pretraining data of Large Language Models and measure their impacts on LLMs using benchmarks about nine major categories. |
| Outcome: | The proposed analysis provides insights into the organization of data to support more efficient pretraining of Large Language Models. |
Syntactic Data Augmentation Increases Robustness to Inference Heuristics (2020.acl-main)
Copied to clipboard
| Challenge: | Pretrained neural models lack sensitivity to word order on controlled challenge sets . augmentation methods that improve accuracy on standard training sets may be a problem . |
| Approach: | They propose to augment standard training sets with syntactically informative examples by applying syntastic transformations to sentences from the MNLI corpus. |
| Outcome: | The proposed method improved BERT’s accuracy on controlled examples that diagnose sensitivity to word order from 0.28 to 0.73 without affecting performance on the MNLI test set. |
An Empirical Study on Robustness to Spurious Correlations using Pre-trained Language Models (2020.tacl-1)
Copied to clipboard
| Challenge: | Recent work shows that pre-trained language models perform poorly on challenging datasets where spurious correlations do not hold. |
| Approach: | They propose to use multi-task learning to improve generalization from minority examples . they propose to combine MTL with auxiliary tasks to improve performance . |
| Outcome: | The proposed model generalizes from minority examples without hurting in-distribution performance. |