Token Encoding for Semantic Recovery
Fuente:
arXiv
Gespeichert in:
| Hauptverfasser: | , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2026
|
| Schlagworte: | |
| Online-Zugang: | |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866914472493318144 |
|---|---|
| author | Hu, Jingzhi Li, Geoffrey Ye |
| author_facet | Hu, Jingzhi Li, Geoffrey Ye |
| contents | Token-based semantic communication is promising for future wireless networks, as it can compact semantic tokens under very limited channel capacity. However, harsh wireless channels often cause missing tokens, leading to severe distortion that prevents reliable semantic recovery at the receiver. In this article, we propose a token encoding framework for robust semantic recovery (TokCode), which incurs no additional transmission overhead and supports plug-and-play deployment. For efficient token encoder optimization, we develop a sentence-semantic-guided foundation model adaptation algorithm (SFMA) that avoids costly end-to-end training. Based on simulation results on prompt-based generative image transmission, TokCode mitigates semantic distortion and can approach the performance upper-bound, even under harsh channels where 40% to 60% of tokens are randomly lost. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2604_12931 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | Token Encoding for Semantic Recovery Hu, Jingzhi Li, Geoffrey Ye Signal Processing Machine Learning Token-based semantic communication is promising for future wireless networks, as it can compact semantic tokens under very limited channel capacity. However, harsh wireless channels often cause missing tokens, leading to severe distortion that prevents reliable semantic recovery at the receiver. In this article, we propose a token encoding framework for robust semantic recovery (TokCode), which incurs no additional transmission overhead and supports plug-and-play deployment. For efficient token encoder optimization, we develop a sentence-semantic-guided foundation model adaptation algorithm (SFMA) that avoids costly end-to-end training. Based on simulation results on prompt-based generative image transmission, TokCode mitigates semantic distortion and can approach the performance upper-bound, even under harsh channels where 40% to 60% of tokens are randomly lost. |
| title | Token Encoding for Semantic Recovery |
| topic | Signal Processing Machine Learning |
| url | https://arxiv.org/abs/2604.12931 |