SERC: LDPC-Inspired Semantic Error Correction for Retrieval-Augmented Generation

Researchers have developed SERC, a semantic error correction framework inspired by LDPC codes that treats LLM text generation as a noisy communication channel to detect and fix hallucinations. The training-free, model-agnostic approach uses sparse verification queries against external evidence, outperforming existing self-correction and retrieval-augmented methods on factual precision benchmarks. SERC enables smaller language models to surpass larger baselines in hallucination reduction while reducing verification overhead, offering a cost-effective solution for resource-constrained environments.

arXiv:2605.28837v1 Announce Type: new Abstract: While Large Language Models LLMs have demonstrated remarkable capabilities, their reliability is significantly compromised by hallucinations. Existing intrinsic self-correction methods attempt to address this, but often fail due to self-bias, where models struggle to identify errors in their own outputs without external verification. To overcome these limitations, we propose the LDPC-inspired semantic error correction for retrieval-augmented generation SERC , providing a theoretical framework to interpret and mitigate LLM hallucinations. We reformulate the text generation process as a semantic noisy channel, treating generated responses as noise-corrupted codewords. Inspired by low-density parity-check LDPC codes, SERC employs a sparse verification strategy: instead of exhaustively checking all facts, it generates low-density verification queries and validates them against external evidence to efficiently detect and correct errors. We evaluate SERC on LongForm Bio and TruthfulQA benchmarks using Llama-3-8B and Qwen2.5-14B. Experimental results demonstrate that SERC outperforms both intrinsic self-correction methods and strong retrieval-augmented baselines, demonstrating significant gains especially in factual precision FactScore . Notably, SERC enables small language models SLMs to surpass the performance of larger baselines in hallucination reduction and information preservation. Our findings demonstrate that SERC provides a training-free, model-agnostic solution that significantly reduces verification overhead compared to dense methods, achieving an optimal trade-off between cost and fidelity in resource-constrained environments.