LLM-Viterbi: Semantic-Aware Decoding for Convolutional Codes
Fuente:
arXiv
Guardado en:
| Autores principales: | , , , , |
|---|---|
| Formato: | Preprint |
| Publicado: |
2026
|
| Materias: | |
| Acceso en línea: | |
| Etiquetas: |
Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
|
| _version_ | 1866918459098529792 |
|---|---|
| author | Li, Zhengtong Yue, Chentao Hao, Jiafu Vucetic, Branka Li, Yonghui |
| author_facet | Li, Zhengtong Yue, Chentao Hao, Jiafu Vucetic, Branka Li, Yonghui |
| contents | Traditional wireless communications rely solely on bit-level channel coding for error correction, without exploiting the inherent linguistic structure of the data source. This paper proposes a large language model (LLM) Viterbi decoder that integrates LLM priors into the Viterbi decoding for text transmission over AWGN channels. The proposed decoder maintains multiple candidate paths during the Viterbi decoding and periodically evaluates path reliabilities using a fine-tuned Byte-level T5 (ByT5) language model. By combining channel reliability metrics with semantic probability from the LLM, it outputs the path that maximizes the joint likelihood of channel observations and linguistic coherence. Simulations show that our decoder achieves significant performance gains over conventional Viterbi decoding in terms of both block error rate (BLER) and semantic similarity. For convolutional codes with constraint length 3, it achieves approximately 1.5 dB more coding gain in BLER, with over 50% improvements in semantic similarity. The framework can extend to other structured data sources beyond text. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2604_19035 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | LLM-Viterbi: Semantic-Aware Decoding for Convolutional Codes Li, Zhengtong Yue, Chentao Hao, Jiafu Vucetic, Branka Li, Yonghui Information Theory 94B10, 68T50 E.4; I.2.7 Traditional wireless communications rely solely on bit-level channel coding for error correction, without exploiting the inherent linguistic structure of the data source. This paper proposes a large language model (LLM) Viterbi decoder that integrates LLM priors into the Viterbi decoding for text transmission over AWGN channels. The proposed decoder maintains multiple candidate paths during the Viterbi decoding and periodically evaluates path reliabilities using a fine-tuned Byte-level T5 (ByT5) language model. By combining channel reliability metrics with semantic probability from the LLM, it outputs the path that maximizes the joint likelihood of channel observations and linguistic coherence. Simulations show that our decoder achieves significant performance gains over conventional Viterbi decoding in terms of both block error rate (BLER) and semantic similarity. For convolutional codes with constraint length 3, it achieves approximately 1.5 dB more coding gain in BLER, with over 50% improvements in semantic similarity. The framework can extend to other structured data sources beyond text. |
| title | LLM-Viterbi: Semantic-Aware Decoding for Convolutional Codes |
| topic | Information Theory 94B10, 68T50 E.4; I.2.7 |
| url | https://arxiv.org/abs/2604.19035 |