Mitigating the Diminishing Effect of Elastic Weight Consolidation (2022.coling-1)
Copied to clipboard
| Challenge: | Existing work addresses catastrophic forgetting in sequential training by fine-tuning pre-trained language models on different datasets. |
| Approach: | They propose to rescale the components of EWC to mitigate catastrophic forgetting by mixing new and old training data and retraining the model from scratch. |
| Outcome: | The proposed method requires smaller values for the trade-off parameters to achieve comparable results to EWC on natural language inference and fact-checking tasks. |
Similar Papers
Elastic weight consolidation for better bias inoculation (2021.eacl-main)
Copied to clipboard
| Challenge: | Recent studies have shown that the lack of suitable inductive biases in sentence-pair classification models can cause misclassifications on training datasets. |
| Approach: | They propose to use elastic weight consolidation (EWC) to fine-tune models to mitigate biases while being less susceptible to catastrophic forgetting. |
| Outcome: | The proposed model improves on fact verification and stress tests while maintaining the original task accuracy. |
Overcoming Catastrophic Forgetting During Domain Adaptation of Neural Machine Translation (N19-1)
Copied to clipboard
| Challenge: | Neural Machine Translation (NMT) performs poorly without large training corpora. |
| Approach: | They propose a machine learning method that retains the majority of general-domain performance lost in continued training without degrading in-domain. |
| Outcome: | The proposed method retains the majority of general-domain performance lost in continued training without degrading in-domain performances. |
An Empirical Investigation Towards Efficient Multi-Domain Language Model Pre-training (2020.emnlp-main)
Copied to clipboard
| Challenge: | Pre-training large language models is a standard practice in the natural language processing community. |
| Approach: | They propose to use elastic weight consolidation to mitigate catastrophic forgetting when pre-trained large language models are evaluated on generic benchmarks. |
| Outcome: | The proposed model achieves state-of-the-art on out-of domain tasks with minimal pre-training . elastic weight consolidation provides best overall scores yielding only a 0.33% drop in performance across seven generic tasks while remaining competitive in bio-medical tasks. |
Unsupervised Pretraining for Neural Machine Translation Using Elastic Weight Consolidation (P19-2)
Copied to clipboard
| Challenge: | Neural machine translation (NMT) uses sequence to sequence architectures, but requires a huge amount of parallel data. |
| Approach: | They use Elastic Weight Consolidation to regularize weights of two language models . they then fine-tune the model on parallel data to avoid forgetting the original task . |
| Outcome: | The proposed method achieves BLEU scores similar to the previous work, but is slower and requires less training data. |
Training Dynamics Underlying Language Model Scaling Laws: Loss Deceleration and Zero-Sum Learning (2025.acl-long)
Copied to clipboard
| Challenge: | Increasing language model size improves cross-entropy loss with power-law behaviour, but scaling laws do not explain how scaling improves loss. |
| Approach: | They find that language models undergo loss deceleration early in training . they attribute loss deceleration to a type of degenerate training dynamics we call zero-sum learning . |
| Outcome: | The proposed scaling improves loss on language models, but degrades loss in other subsets, resulting in bottlenecks. |
Pitfalls of Scale: Investigating the Inverse Task of Redefinition in Large Language Models (2025.findings-acl)
Copied to clipboard
| Challenge: | Large Language Models (LLMs) have shown remarkable results in several linguistic, reasoning and knowledge retrieval tasks. |
| Approach: | They propose to scale Large Language Models (LLMs) to scale up to reveal potential reasoning gaps as LLMs scale up. |
| Outcome: | The proposed redefinition task shows that model performance degrades with scale, and false confidence rises. |
Language Models Resist Alignment: Evidence From Data Compression (2025.acl-long)
Copied to clipboard
Jiaming Ji, Kaile Wang, Tianyi Alex Qiu, Boyuan Chen, Jiayi Zhou, Changye Li, Hantao Lou, Josef Dai, Yunhuai Liu, Yaodong Yang
| Challenge: | Large language models (LLMs) may exhibit undesirable behaviors due to the inevitable biases and harmful content present in training. |
| Approach: | They propose to investigate the elasticity of large language models by examining their performance. |
| Outcome: | The proposed model performance declines rapidly before reverting to the pre-training distribution, the authors show . the proposed model weight and code are available at pku-lm-res ist-alignment.github.io. |
Revisiting Catastrophic Forgetting in Large Language Model Tuning (2024.findings-emnlp)
Copied to clipboard
| Challenge: | Catastrophic Forgetting (CF) compromises the effectiveness of large language models during fine-tuning, yet the underlying causes of CF remain largely unexplored. |
| Approach: | They propose a method to flatten the model loss landscape to mitigate CF by flattening the loss landscape. |
| Outcome: | The proposed method complements existing anti-forgetting strategies, further enhancing the resistance of LLMs to CF. |
Improving Scheduled Sampling with Elastic Weight Consolidation for Neural Machine Translation (2022.findings-emnlp)
Copied to clipboard
| Challenge: | Autoregressive models trained with maximum likelihood estimation suffer from exposure bias, i.e. the discrepancy between ground-truth prefixes used during training and model-generated prefix at inference time. |
| Approach: | They propose to use Elastic Weight Consolidation to better balance mitigating exposure bias with retaining performance. |
| Outcome: | The proposed method significantly outperforms maximum likelihood estimation and scheduled sampling baselines on four translation datasets. |
Emergent Inabilities? Inverse Scaling Over the Course of Pretraining (2023.findings-emnlp)
Copied to clipboard
| Challenge: | Recent research has found that increased number of model parameters and increased size of the training dataset positively influence model performance. |
| Approach: | They investigate whether language models' performance on specific tasks can decrease over the course of training. |
| Outcome: | The proposed model size-based scaling is found on 8 tasks on which Pythia 12B shows decreased performance over the course of training. |