Saved in:
| Main Authors: | , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2506.05672 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866910991758917632 |
|---|---|
| author | Zhmoginov, Andrey Lee, Jihwan Vladymyrov, Max Sandler, Mark |
| author_facet | Zhmoginov, Andrey Lee, Jihwan Vladymyrov, Max Sandler, Mark |
| contents | Large Language Models (LLMs) based on Transformers excel at text processing, but their reliance on prompts for specialized behavior introduces computational overhead. We propose a modification to a Transformer architecture that eliminates the need for explicit prompts by learning to encode context into the model's weights. Our Contextually Guided Transformer (CGT) model maintains a contextual summary at each sequence position, allowing it to update the weights on the fly based on the preceding context. This approach enables the model to self-specialize, effectively creating a tailored model for processing information following a given prefix. We demonstrate the effectiveness of our method on synthetic in-context learning tasks and language modeling benchmarks. Furthermore, we introduce techniques for enhancing the interpretability of the learned contextual representations, drawing connections to Variational Autoencoders and promoting smoother, more consistent context encoding. This work offers a novel direction for efficient and adaptable language modeling by integrating context directly into the model's architecture. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2506_05672 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Contextually Guided Transformers via Low-Rank Adaptation Zhmoginov, Andrey Lee, Jihwan Vladymyrov, Max Sandler, Mark Machine Learning Computation and Language Large Language Models (LLMs) based on Transformers excel at text processing, but their reliance on prompts for specialized behavior introduces computational overhead. We propose a modification to a Transformer architecture that eliminates the need for explicit prompts by learning to encode context into the model's weights. Our Contextually Guided Transformer (CGT) model maintains a contextual summary at each sequence position, allowing it to update the weights on the fly based on the preceding context. This approach enables the model to self-specialize, effectively creating a tailored model for processing information following a given prefix. We demonstrate the effectiveness of our method on synthetic in-context learning tasks and language modeling benchmarks. Furthermore, we introduce techniques for enhancing the interpretability of the learned contextual representations, drawing connections to Variational Autoencoders and promoting smoother, more consistent context encoding. This work offers a novel direction for efficient and adaptable language modeling by integrating context directly into the model's architecture. |
| title | Contextually Guided Transformers via Low-Rank Adaptation |
| topic | Machine Learning Computation and Language |
| url | https://arxiv.org/abs/2506.05672 |