Real-time Verification and Refinement of Language Model Text Generation

2025-01-14

AI Summary

Early factual errors in large language model (LLM) generation propagate cascading inaccuracies, while conventional post-hoc verification suffers from high latency and low efficiency. Method: This paper proposes Streaming-VR, a streaming validation and correction framework featuring a novel token-level incremental verification mechanism. It concurrently invokes lightweight auxiliary models to perform fine-grained factual checking on each generated token segment during autoregressive LLM decoding and dynamically triggers localized regeneration upon error detection. The framework leverages a multi-model collaborative architecture and streaming-aware prompt engineering to overcome temporal bottlenecks inherent in retrospective correction. Contribution/Results: Experiments demonstrate that Streaming-VR significantly improves answer accuracy across multiple factuality benchmarks, reduces verification latency by 47%, and achieves superior end-to-end response efficiency compared to state-of-the-art approaches.

Technology Category

• Natural Language Processing: Fact-Checking / Misinformation Detection (NLP Focus)
• Machine Learning: Large Multimodal Models (LMMs)
• Computer Vision: Large Vision Models

Application Category

• Search and Retrieval-Augmented AI: Large language models for search
• User Modeling, Personalization and Recommendation: Large Language Models (LLM) for user modeling and recommendation
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs

Abstract

Large language models (LLMs) have shown remarkable performance across a wide range of natural language tasks. However, a critical challenge remains in that they sometimes generate factually incorrect answers. To address this, while many previous work has focused on identifying errors in their generation and further refining them, they are slow in deployment since they are designed to verify the response from LLMs only after their entire generation (from the first to last tokens) is done. Further, we observe that once LLMs generate incorrect tokens early on, there is a higher likelihood that subsequent tokens will also be factually incorrect. To this end, in this work, we propose Streaming-VR (Streaming Verification and Refinement), a novel approach designed to enhance the efficiency of verification and refinement of LLM outputs. Specifically, the proposed Streaming-VR enables on-the-fly verification and correction of tokens as they are being generated, similar to a streaming process, ensuring that each subset of tokens is checked and refined in real-time by another LLM as the LLM constructs its response. Through comprehensive evaluations on multiple datasets, we demonstrate that our approach not only enhances the factual accuracy of LLMs, but also offers a more efficient solution compared to prior refinement methods.

Problem

• Large Language Models
• Error Detection
• Efficiency

Innovation

• Streaming-VR
• Real-time Error Detection
• Large Language Model Optimization