Retrieval-augmented Decoding for Improving Truthfulness in Open-ended Generation

arXiv:2508.02184v3 Announce Type: replace Abstract: Ensuring truthfulness in large language models (LLMs) remains a critical challenge for reliable text generation. While supervised fine-tuning and reinforcement learning with human feedback have shown promise, they require a substantial amount of annotated data and computational resources, limiting scalability. In contrast, decoding-time interventions offer lightweight alternatives without model retraining. However, existing decoding strategies often face issues like prompt sensitivity, limited generalization, or dependence on internal model states. We propose Retrieval-Augmented Decoding (RAD), a context-aware adaptive decoding method that leverages a compact reference grounding space built from as few as 10 annotated examples and comprising pairs of context embeddings and next-token logits from truthful responses, to enable retrieval-based logit shaping during inference. At each decoding step, RAD retrieves high-quality semantically similar contexts from this grounding space and aggregates their associated next token logits to modify the model's current logits. Across four open-ended generation benchmarks and four LLMs, our method consistently outperforms strong baselines and shows robust cross-task generalization, underscoring the promise of context-aware decoding for enhancing factual reliability.

Abstract:Ensuring truthfulness in large language models (LLMs) remains a critical challenge for reliable text generation. While supervised fine-tuning and reinforcement learning with human feedback have shown promise, they require a substantial amount of annotated data and computational resources, limiting scalability. In contrast, decoding-time interventions offer lightweight alternatives without model retraining. However, existing decoding strategies often face issues like prompt sensitivity, limited generalization, or dependence on internal model states. We propose Retrieval-Augmented Decoding (RAD), a context-aware adaptive decoding method that leverages a compact reference grounding space built from as few as 10 annotated examples and comprising pairs of context embeddings and next-token logits from truthful responses, to enable retrieval-based logit shaping during inference. At each decoding step, RAD retrieves high-quality semantically similar contexts from this grounding space and aggregates their associated next token logits to modify the model's current logits. Across four open-ended generation benchmarks and four LLMs, our method consistently outperforms strong baselines and shows robust cross-task generalization, underscoring the promise of context-aware decoding for enhancing factual reliability.

Machine Learning (cs.LG)

arXiv:2508.02184 [cs.LG]

(or arXiv:2508.02184v3 [cs.LG] for this version)

https://doi.org/10.48550/arXiv.2508.02184 Focus to learn more arXiv-issued DOI via DataCite

https://doi.org/10.1007/978-3-032-37673-2_2 Focus to learn more DOI(s) linking to related resources

arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.

Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.

Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.

aiqbench analysis

Retrieval-augmented Decoding for Improving Truthfulness in Open-ended Generation

Related coverage