Factors That Support Grounded Responses in LLM Conversations: A Rapid Review

TL;DR

This paper reviews techniques to improve the reliability of LLM responses by aligning outputs with user intent, grounding in context, and reducing hallucinations, focusing on inference-time methods for efficiency.

Contribution

It systematically categorizes and analyzes various alignment strategies across the LLM lifecycle, highlighting inference-time approaches as particularly effective.

Findings

Inference-time methods are efficient and effective.

Structured mechanisms improve response quality.

Techniques support grounding and hallucination mitigation.

Abstract

Large language models (LLMs) may generate outputs that are misaligned with user intent, lack contextual grounding, or exhibit hallucinations during conversation, which compromises the reliability of LLM-based applications. This review aimed to identify and analyze techniques that align LLM responses with conversational goals, ensure grounding, and reduce hallucination and topic drift. We conducted a Rapid Review guided by the PRISMA framework and the PICO strategy to structure the search, filtering, and selection processes. The alignment strategies identified were categorized according to the LLM lifecycle phase in which they operate: inference-time, post-training, and reinforcement learning-based methods. Among these, inference-time approaches emerged as particularly efficient, aligning outputs without retraining while supporting user intent, contextual grounding, and hallucination mitigation. The reviewed techniques provided structured mechanisms for improving the quality and reliability of LLM responses across key alignment objectives.