Knowledge-Aware Self-Correction in Language Models via Structured Memory Graphs
This addresses the issue of hallucinations in LLMs for users needing reliable outputs, but it is incremental as it builds on existing methods without retraining.
The paper tackles the problem of factual errors in large language models by introducing a lightweight framework for self-correction using structured memory graphs, showing promising results on simple factual prompts.
Large Language Models (LLMs) are powerful yet prone to generating factual errors, commonly referred to as hallucinations. We present a lightweight, interpretable framework for knowledge-aware self-correction of LLM outputs using structured memory graphs based on RDF triples. Without retraining or fine-tuning, our method post-processes model outputs and corrects factual inconsistencies via external semantic memory. We demonstrate the approach using DistilGPT-2 and show promising results on simple factual prompts.