Lower Layer Matters: Alleviating Hallucination via Multi-Layer Fusion Contrastive Decoding with Truthfulness Refocused

2024-08-16

arXiv.org

AI Summary

To address the pervasive hallucination problem in large language model (LLM) generation, this paper proposes LOL, a multi-layer fused contrastive decoding framework. LOL introduces latent representations into contrastive decoding for the first time, performing feature-level contrast between the original LLM and a lightweight hallucination-inducing model across both the final layer and multiple intermediate layers. It further incorporates a context-aware veracity refocusing module that dynamically enhances factual semantic encoding. By integrating latent-layer concatenation, contrastive learning, and decoding-time intervention, LOL enables fine-grained modeling of factual consistency. On the TruthfulQA benchmark, LOL achieves an average improvement of 4.5 points over baseline methods—outperforming existing contrastive decoding approaches significantly—while preserving generation quality and effectively mitigating hallucinations.

Technology Category

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

Application Category

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

Abstract

Large Language Models (LLMs) have demonstrated exceptional performance across various natural language processing tasks, yet they occasionally tend to yield content that factually inaccurate or discordant with the expected output, a phenomenon empirically referred to as"hallucination". To tackle this issue, recent works have investigated contrastive decoding between the original model and an amateur model with induced hallucination, which has shown promising results. Nonetheless, this method may undermine the output distribution of the original LLM caused by its coarse contrast and simplistic subtraction operation, potentially leading to errors in certain cases. In this paper, we introduce a novel contrastive decoding framework termed LOL (LOwer Layer Matters). Our approach involves concatenating the contrastive decoding of both the final and lower layers between the original model and the amateur model, thereby achieving multi-layer fusion to aid in the mitigation of hallucination. Additionally, we incorporate a truthfulness refocused module that leverages contextual guidance to enhance factual encoding, further capturing truthfulness during contrastive decoding. Extensive experiments conducted on two publicly available datasets illustrate that our proposed LOL framework can substantially alleviate hallucination while surpassing existing baselines in most cases. Compared with the best baseline, we improve by average 4.5 points on all metrics of TruthfulQA. The source code is coming soon.

Problem

• Mitigating hallucinations in Large Language Models (LLMs)
• Improving truthfulness via multi-layer fusion contrastive decoding
• Addressing output distribution disruption in contrastive decoding methods

Innovation

• Multi-layer fusion contrastive decoding
• Truthfulness refocused module
• Lower layers integration for hallucination reduction