# Improving Factuality in Large Language Models via Decoding-Time Hallucinatory and Truthful Comparators

arXiv: https://arxiv.org/abs/2408.12325 · HTML: https://mlsanity.com/paper/2408.12325/

## Key facts

- Cited by 19 papers in the map (data through 14 Oct 2026).
- Builds on 38 of its 55 references that are in the map.
- First posted on arXiv 22 Aug 2024.

## Builds on (most cited first, 30 of 38)

- [Training language models to follow instructions with human feedback](https://mlsanity.com/paper/2203.02155/) (2022)
- [LoRA: Low-Rank Adaptation of Large Language Models](https://mlsanity.com/paper/2106.09685/) (2021)
- [Llama 2: Open Foundation and Fine-Tuned Chat Models](https://mlsanity.com/paper/2307.09288/) (2023)
- [Direct Preference Optimization: Your Language Model is Secretly a Reward Model](https://mlsanity.com/paper/2305.18290/) (2023)
- [Towards Deep Learning Models Resistant to Adversarial Attacks](https://mlsanity.com/paper/1706.06083/) (2017)
- [BERTScore: Evaluating Text Generation with BERT](https://mlsanity.com/paper/1904.09675/) (2019)
- [Training a Helpful and Harmless Assistant with Reinforcement Learning from Human Feedback](https://mlsanity.com/paper/2204.05862/) (2022)
- [TruthfulQA: Measuring How Models Mimic Human Falsehoods](https://mlsanity.com/paper/2109.07958/) (2021)
- [Representation Engineering: A Top-Down Approach to AI Transparency](https://mlsanity.com/paper/2310.01405/) (2023)
- [Don't Give Me the Details, Just the Summary! Topic-Aware Convolutional Neural Networks for Extreme Summarization](https://mlsanity.com/paper/1808.08745/) (2018)
- [Inference-Time Intervention: Eliciting Truthful Answers from a Language Model](https://mlsanity.com/paper/2306.03341/) (2023)
- [Get To The Point: Summarization with Pointer-Generator Networks](https://mlsanity.com/paper/1704.04368/) (2017)
- [LIMA: Less Is More for Alignment](https://mlsanity.com/paper/2305.11206/) (2023)
- [A Multitask, Multilingual, Multimodal Evaluation of ChatGPT on Reasoning, Hallucination, and Interactivity](https://mlsanity.com/paper/2302.04023/) (2023)
- [Discovering Latent Knowledge in Language Models Without Supervision](https://mlsanity.com/paper/2212.03827/) (2022)
- [Baichuan 2: Open Large-scale Language Models](https://mlsanity.com/paper/2309.10305/) (2023)
- [Textbooks Are All You Need II: phi-1.5 technical report](https://mlsanity.com/paper/2309.05463/) (2023)
- [DExperts: Decoding-Time Controlled Text Generation with Experts and Anti-Experts](https://mlsanity.com/paper/2105.03023/) (2021)
- [DoLa: Decoding by Contrasting Layers Improves Factuality in Large Language Models](https://mlsanity.com/paper/2309.03883/) (2023)
- [Check Your Facts and Try Again: Improving Large Language Models with External Knowledge and Automated Feedback](https://mlsanity.com/paper/2302.12813/) (2023)
- [Adversarial Training Methods for Semi-Supervised Text Classification](https://mlsanity.com/paper/1605.07725/) (2016)
- [FacTool: Factuality Detection in Generative AI -- A Tool Augmented Framework for Multi-Task and Multi-Domain Scenarios](https://mlsanity.com/paper/2307.13528/) (2023)
- [Fine-tuning Language Models for Factuality](https://mlsanity.com/paper/2311.08401/) (2023)
- [Chain-of-Thought Reasoning Without Prompting](https://mlsanity.com/paper/2402.10200/) (2024)
- [Survey on Factuality in Large Language Models: Knowledge, Retrieval and Domain-Specificity](https://mlsanity.com/paper/2310.07521/) (2023)
- [Evaluating Factuality in Generation with Dependency-level Entailment](https://mlsanity.com/paper/2010.05478/) (2020)
- [Mixture of Cluster-conditional LoRA Experts for Vision-language Instruction Tuning](https://mlsanity.com/paper/2312.12379/) (2023)
- [TruthX: Alleviating Hallucinations by Editing Large Language Models in Truthful Space](https://mlsanity.com/paper/2402.17811/) (2024)
- [Alleviating Hallucinations of Large Language Models through Induced Hallucinations](https://mlsanity.com/paper/2312.15710/) (2023)
- [FactKB: Generalizable Factuality Evaluation using Language Models Enhanced with Factual Knowledge](https://mlsanity.com/paper/2305.08281/) (2023)

## Built on by (most cited first, 19 of 19)

- [Critical Tokens Matter: Token-Level Contrastive Estimation Enhances LLM's Reasoning Capability](https://mlsanity.com/paper/2411.19943/) (2024)
- [Knowledge Boundary of Large Language Models: A Survey](https://mlsanity.com/paper/2412.12472/) (2024)
- [Reinforcement Learning Meets Large Language Models: A Survey of Advancements and Applications Across the LLM Lifecycle](https://mlsanity.com/paper/2509.16679/) (2025)
- [Resolving Evidence Sparsity: Agentic Context Engineering for Long-Document Understanding](https://mlsanity.com/paper/2511.22850/) (2025)
- [Privacy-Aware Decoding: Mitigating Privacy Leakage of Large Language Models in Retrieval-Augmented Generation](https://mlsanity.com/paper/2508.03098/) (2025)
- [Loki's Dance of Illusions: A Comprehensive Survey of Hallucination in Large Language Models](https://mlsanity.com/paper/2507.02870/) (2025)
- [SAILViT: Towards Robust and Generalizable Visual Backbones for MLLMs via Gradual Feature Refinement](https://mlsanity.com/paper/2507.01643/) (2025)
- [AdaLRS: Loss-Guided Adaptive Learning Rate Search for Efficient Foundation Model Pretraining](https://mlsanity.com/paper/2506.13274/) (2025)
- [The Illusionist's Prompt: Exposing the Factual Vulnerabilities of Large Language Models with Linguistic Nuances](https://mlsanity.com/paper/2504.02865/) (2025)
- [Knowledge Injection Exists in MoE? Exploring Expert-Aware Contrast Decoding in MoE for Mitigating LLMs'Hallucinations](https://mlsanity.com/paper/2607.20426/) (2026)
- [Noise Augmented Fine Tuning for Mitigating Hallucinations in Large Language Models](https://mlsanity.com/paper/2504.03302/) (2025)
- [Expanding before Inferring: Enhancing Factuality in Large Language Models through Premature Layers Interpolation](https://mlsanity.com/paper/2506.02973/) (2025)
- [Context-Fidelity Boosting: Enhancing Faithful Generation through Watermark-Inspired Decoding](https://mlsanity.com/paper/2604.22335/) (2026)
- [Beyond Tokens: A Survey on Decoding Methods for Large Language and Vision-Language Models](https://mlsanity.com/paper/2608.14797/) (2026)
- [SUP-MIMIC: A Multi-Task Clinical Diagnosis Benchmark for Evaluating LLMs' Robustness to Contradictory Evidence](https://mlsanity.com/paper/2608.29582/) (2026)
- [Look Before You Leap: Factual Decoding with Internal Attribution Signals](https://mlsanity.com/paper/2609.15745/) (2026)
- [OmniFysics-Captioner Technical Report: Grounding Omni-Modal Understanding in the Physical World for Better Captioning](https://mlsanity.com/paper/2609.31714/) (2026)
- [Fysiverse-3D-SimReady Technical Report: Agentic Physical Simulation for Pragmatic 3D World Reconstruction](https://mlsanity.com/paper/2609.31715/) (2026)
- [Beam Search as Test-Time Self-Distillation via Counterfactual Contexts](https://mlsanity.com/paper/2609.37041/) (2026)

## Abstract (arXiv)

Despite their remarkable capabilities, Large Language Models (LLMs) are prone to generate responses that contradict verifiable facts, i.e., unfaithful hallucination content. Existing efforts generally focus on optimizing model parameters or editing semantic representations, which compromise the internal factual knowledge of target LLMs. In addition, hallucinations typically exhibit multifaceted patterns in downstream tasks, limiting the model's holistic performance across tasks. In this paper, we propose a Comparator-driven Decoding-Time (CDT) framework to alleviate the response hallucination. Firstly, we construct hallucinatory and truthful comparators with multi-task fine-tuning samples. In this case, we present an instruction prototype-guided mixture of experts strategy to enhance the ability of the corresponding comparators to capture different hallucination or truthfulness patterns in distinct task instructions. CDT constrains next-token predictions to factuality-robust distributions by contrasting the logit differences between the target LLMs and these comparators. Systematic experiments on multiple downstream tasks show that our framework can significantly improve the model performance and response factuality.