Generative Semantic HARQ: Latent-Space Text Retransmission and Combining

2026-03-16

AI Summary

This work proposes a semantic-level hybrid automatic repeat request (HARQ) framework to address the insufficient reliability of semantic transmission in semantic communication systems. Built upon the conventional protocol stack, the framework incorporates a lightweight Transformer-VAE codec to enable semantic retransmission and fusion in the latent space. It innovatively employs a stochastic encoder to generate diverse latent representations, thereby providing incremental semantic information without incurring additional protocol overhead. Furthermore, a self-consistency-based soft quality estimator and a quality-aware fusion mechanism are designed to enhance semantic fidelity. Under mixed semantic distortion conditions, the integration of weighted averaging or maximal ratio combining (MRC)-inspired fusion strategies with a self-consistency-driven HARQ triggering mechanism significantly improves the reliability of semantic communication.

Technology Category

• Multiagent Systems: Agent Communication
• Cognitive Modeling & Cognitive Systems: Neural Spike Coding
• Search and Optimization: Sampling/Simulation-based Search

Application Category

• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Search and Retrieval-Augmented AI: Vertical and domain-specific search
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications

Abstract

Semantic communication conveys meaning rather than raw bits, but reliability at the semantic level remains an open challenge. We propose a semantic-level hybrid automatic repeat request (HARQ) framework for text communication, in which a Transformer-variational autoencoder (VAE) codec operates as a lightweight overlay on the conventional protocol stack. The stochastic encoder inherently generates diverse latent representations across retransmissions-providing incremental knowledge (IK) from a single model without dedicated protocol design. On the receiver side, a soft quality estimator triggers retransmissions and a quality-aware combiner merges the received latent vectors within a consistent latent space. We systematically benchmark six semantic quality metrics and four soft combining strategies under hybrid semantic distortion that mixes systematic bias with additive noise. The results suggest combining Weighted-Average or MRC-Inspired combining with self-consistency-based HARQ triggering for the best performance.

Problem

• Semantic Communication
• HARQ
• Latent Space
• Text Retransmission
• Reliability

Innovation

• Semantic HARQ
• Latent-space retransmission
• Incremental knowledge
• Quality-aware combining
• Transformer-VAE