Too Much in Common: Shifting of Embeddings in Transformer Language Models and its Implications (2021.naacl-main)
Copied to clipboard
| Challenge: | Existing studies have shown that word embeddings do not occupy a narrow cone, but rather drift in common directions. |
| Approach: | They show that anisotropy can be restored using a simple transformation of word embeddings. |
| Outcome: | The proposed model can restore anisotropy using a simple transformation. |
Similar Papers
Anisotropy is Not Inherent to Transformers (2024.naacl-long)
Copied to clipboard
| Challenge: | Existing work has shown that Transformer embeddings are anisotropic, which is called the representation degradation problem. |
| Approach: | They identify a set of Transformer models with isotropic embedding spaces, the large Pythia models. |
| Outcome: | The proposed model sets show that isotropic models do not develop as previously theorized. |
How to Dissect a Muppet: The Structure of Transformer Embedding Spaces (2022.tacl-1)
Copied to clipboard
| Challenge: | Pretrained embeddings based on the Transformer architecture have taken the NLP community by storm . a novel decomposition of Transformer output embeddables is demonstrated . |
| Approach: | They propose to decompose Transformer output embeddings into a sum of vector factors . they show multi-head attentions and feed-forwards are not equally useful in downstream applications . |
| Outcome: | The proposed method outperforms recurrent architectures on a wide variety of tasks. |
Do Transformer Modifications Transfer Across Implementations and Applications? (2021.emnlp-main)
Copied to clipboard
Sharan Narang, Hyung Won Chung, Yi Tay, Liam Fedus, Thibault Fevry, Michael Matena, Karishma Malkan, Noah Fiedel, Noam Shazeer, Zhenzhong Lan, Yanqi Zhou, Wei Li, Nan Ding, Jake Marcus, Adam Roberts, Colin Raffel
| Challenge: | Currently, the Transformer is the de facto architecture of choice for processing sequential data. |
| Approach: | They evaluate the Transformer architecture and its modifications in a shared experimental setting . they conjecture that performance improvements may strongly depend on implementation details . |
| Outcome: | The proposed improvements do not significantly improve performance, the authors find . the proposed improvements are either developed in the same codebase or are minor changes . |
Transformer Grammars: Augmenting Transformer Language Models with Syntactic Inductive Biases at Scale (2022.tacl-1)
Copied to clipboard
| Challenge: | a novel class of Transformer language models that combine expressive power, scalability, and strong performance of Transformers and recursive syntactic compositions. |
| Approach: | They introduce Transformer Grammars, a class of Transformer language models that combine expressive power and recursive syntactic compositions. |
| Outcome: | The proposed model outperforms strong baselines on sentence-level language modeling perplexity and syntax-sensitive language evaluation metrics. |
A Meta-Learning Perspective on Transformers for Causal Language Modeling (2024.findings-acl)
Copied to clipboard
| Challenge: | Mechanisms of the Transformer architecture for causal language modeling are not well understood. |
| Approach: | They propose a meta-learning view of the Transformer architecture when trained for a causal language modeling task by explicating an inner optimization process that may happen within the Transformer. |
| Outcome: | The proposed model is based on a self-attention mechanism and has been widely used in natural language processing, computer vision, and scientific discovery. |
The Shape of Learning: Anisotropy and Intrinsic Dimensions in Transformer-Based Models (2024.findings-eacl)
Copied to clipboard
Anton Razzhigaev, Matvey Mikhalchuk, Elizaveta Goncharova, Ivan Oseledets, Denis Dimitrov, Andrey Kuznetsov
| Challenge: | Embeddings in transformers encode vast amounts of linguistic nuances and patterns. |
| Approach: | They investigate the anisotropy dynamics and intrinsic dimension of embeddings in transformers . they found that transformer decoders exhibit a bell-shaped anisotropie profile . |
| Outcome: | The investigated embeddings exhibit a bell-shaped curve with the highest anisotropy concentrations in the middle layers . the intrinsic dimension increases in the initial phases of training, indicating an expansion into higher-dimensional space. |
Latent Positional Information is in the Self-Attention Variance of Transformer Language Models Without Positional Embeddings (2023.acl-short)
Copied to clipboard
| Challenge: | Recent research has called into question the necessity of positional embeddings in transformer language models. |
| Approach: | They propose to discard positional embeddings in transformer language models to facilitate more efficient pretraining. |
| Outcome: | The proposed model encodes strong positional information through shrinkage of self-attention variance. |
Beyond Position: the emergence of wavelet-like properties in Transformers (2025.acl-long)
Copied to clipboard
| Challenge: | Despite its widespread adoption, theoretical limitations in positional encodings are resolved by developing emergent, wavelet-like processing strategies. |
| Approach: | They propose to use Rotary Position Embeddings to develop emergent, wavelet-like properties that compensate for the positional encoding’s theoretical limitations. |
| Outcome: | The attention heads evolve to implement multi-resolution processing analogous to wavelet transforms. |
Can Transformers Learn n-gram Language Models? (2024.emnlp-main)
Copied to clipboard
| Challenge: | Existing work has tested transformers' ability to represent formal languages, but language models are not classifiers of strings but rather distributions over them. |
| Approach: | They relate transformers' ability to learn random n-gram language models to ngram language model (LM) they find add- smoothing outperforms transformers on the former, while transformers perform better on the latter . |
| Outcome: | The proposed models outperform classical methods designed to learn n-gram LMs, while transformers perform better on the latter. |
The Hidden Space of Transformer Language Adapters (2024.acl-long)
Copied to clipboard
| Challenge: | Adapters are small modules trained on top of a frozen language model to adapt predictions to new target languages. |
| Approach: | They propose to train transformer language adapters on top of a frozen model to adapt predictions to new target languages. |
| Outcome: | The transformer language adapters are trained on top of a frozen model to adapt predictions to new target languages. |