Reinforcement-Learned Unequal Error Protection for Quantized Semantic Embeddings

2026-01-01

arXiv.org

AI Summary

This work addresses the challenge of preserving both global semantic similarity and critical entity fidelity in bandwidth-constrained semantic communication systems, where conventional channel coding falls short. The authors propose a reinforcement learning–based adaptive repetition coding framework that, for the first time, leverages reinforcement learning for fine-grained semantic protection. By quantifying semantic embeddings and evaluating the importance of individual dimensions, the method dynamically allocates unequal error protection resources. Departing from semantics-agnostic coding paradigms, this approach achieves significant performance gains over traditional schemes using only simple repetition coding. Experimental results demonstrate that, at 1 dB SNR, the proposed method improves chrF score by 6.8% and entity retention rate by 9.3% compared to uniform protection strategies, effectively enhancing semantic communication performance.

Technology Category

• Search and Optimization: Learning to Search
• Natural Language Processing: Sentence-level Semantics, Textual Inference, etc.
• Machine Learning: Semi-Supervised Learning

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: Web learning to rank, online learning, and counterfactual learning for ranking
• User Modeling, Personalization and Recommendation: User privacy protection in personalized systems

Abstract

This paper tackles the pressing challenge of preserving semantic meaning in communication systems constrained by limited bandwidth. We introduce a novel reinforcement learning framework that achieves per-dimension unequal error protection via adaptive repetition coding. Central to our approach is a composite semantic distortion metric that balances global embedding similarity with entity-level preservation, empowering the reinforcement learning agent to allocate protection in a context-aware manner. Experiments show statistically significant gains over uniform protection, achieving 6.8% higher chrF scores and 9.3% better entity preservation at 1 dB SNR. The key innovation of our framework is the demonstration that simple, intelligently allocated repetition coding enables fine-grained semantic protection -- an advantage unattainable with conventional codes such as LDPC or Reed-Solomon. Our findings challenge traditional channel coding paradigms by establishing that code structure must align with semantic granularity. This approach is particularly suited to edge computing and IoT scenarios, where bandwidth is scarce, but semantic fidelity is critical, providing a practical pathway for next-generation semantic-aware networks.

Problem

• semantic communication
• unequal error protection
• quantized embeddings
• bandwidth-constrained
• semantic distortion

Innovation

• reinforcement learning
• unequal error protection
• semantic communication
• adaptive repetition coding
• semantic distortion metric