Token Encoding for Semantic Recovery
TLDR
TokCode enables robust semantic recovery in wireless networks by encoding tokens to mitigate distortion from up to 60% token loss.
Key contributions
- Proposes TokCode, a token encoding framework for robust semantic recovery in wireless channels.
- TokCode incurs no additional transmission overhead and supports plug-and-play deployment.
- Introduces SFMA for efficient token encoder optimization, avoiding costly end-to-end training.
- Mitigates semantic distortion and approaches performance upper-bound even with 40-60% token loss.
Why it matters
Semantic communication is crucial for future wireless networks, but token loss in harsh channels hinders its reliability. This paper introduces TokCode, a novel framework that ensures robust semantic recovery without extra overhead. It significantly improves performance, making semantic communication viable even under severe channel conditions.
Original Abstract
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.
📬 Weekly AI Paper Digest
Get the top 10 AI/ML arXiv papers from the week — summarized, scored, and delivered to your inbox every Monday.