# From Misleading Queries to Accurate Answers: A Three-Stage Fine-Tuning Method for LLMs

arXiv: https://arxiv.org/abs/2504.11277 · HTML: https://mlsanity.com/paper/2504.11277/

## Key facts

- Cited by 2 papers in the map (data through 14 Oct 2026).
- Builds on 21 of its 36 references that are in the map.
- First posted on arXiv 15 Apr 2025.

## Builds on (most cited first, 21 of 21)

- [The Llama 3 Herd of Models](https://mlsanity.com/paper/2407.21783/) (2024)
- [Retrieval-Augmented Generation for Knowledge-Intensive NLP Tasks](https://mlsanity.com/paper/2005.11401/) (2020)
- [Sentence-BERT: Sentence Embeddings using Siamese BERT-Networks](https://mlsanity.com/paper/1908.10084/) (2019)
- [Self-Refine: Iterative Refinement with Self-Feedback](https://mlsanity.com/paper/2303.17651/) (2023)
- [TruthfulQA: Measuring How Models Mimic Human Falsehoods](https://mlsanity.com/paper/2109.07958/) (2021)
- [Retrieval-Augmented Generation for Large Language Models: A Survey](https://mlsanity.com/paper/2312.10997/) (2023)
- [CommonsenseQA: A Question Answering Challenge Targeting Commonsense Knowledge](https://mlsanity.com/paper/1811.00937/) (2018)
- [BLEURT: Learning Robust Metrics for Text Generation](https://mlsanity.com/paper/2004.04696/) (2020)
- [Adversarial NLI: A New Benchmark for Natural Language Understanding](https://mlsanity.com/paper/1910.14599/) (2019)
- [KILT: a Benchmark for Knowledge Intensive Language Tasks](https://mlsanity.com/paper/2009.02252/) (2020)
- [When Can LLMs Actually Correct Their Own Mistakes? A Critical Survey of Self-Correction of LLMs](https://mlsanity.com/paper/2406.01297/) (2024)
- [Can LLM-Generated Misinformation Be Detected?](https://mlsanity.com/paper/2309.13788/) (2023)
- [Modeling Future Conversation Turns to Teach LLMs to Ask Clarifying Questions](https://mlsanity.com/paper/2410.13788/) (2024)
- [CLAM: Selective Clarification for Ambiguous Questions with Generative Language Models](https://mlsanity.com/paper/2212.07769/) (2022)
- [Large Language Models have Intrinsic Self-Correction Ability](https://mlsanity.com/paper/2406.15673/) (2024)
- [Disinformation Detection: An Evolving Challenge in the Age of LLMs](https://mlsanity.com/paper/2309.15847/) (2023)
- [Large Language Models Can Self-Correct with Key Condition Verification](https://mlsanity.com/paper/2405.14092/) (2024)
- [Truth-Aware Context Selection: Mitigating Hallucinations of Large Language Models Being Misled by Untruthful Contexts](https://mlsanity.com/paper/2403.07556/) (2024)
- [Large Language Models are Skeptics: False Negative Problem of Input-conflicting Hallucination](https://mlsanity.com/paper/2406.13929/) (2024)
- [Semantic-aware Contrastive Learning for More Accurate Semantic Parsing](https://mlsanity.com/paper/2301.07919/) (2023)
- [Word Embedding based Edit Distance](https://mlsanity.com/paper/1810.10752/) (2018)

## Built on by (most cited first, 2 of 2)

- [Stable-RAG: Mitigating Retrieval-Permutation-Induced Hallucinations in Retrieval-Augmented Generation](https://mlsanity.com/paper/2601.02993/) (2026)
- [From Passive Response to Proactive Correction: Enhancing LLM Robustness Against Input Fact Perturbations](https://mlsanity.com/paper/2608.25894/) (2026)

## Abstract (arXiv)

Large language models (LLMs) exhibit excellent performance in natural language processing (NLP), but remain highly sensitive to the quality of input queries, especially when these queries contain misleading or inaccurate information. Existing methods focus on correcting the output, but they often overlook the potential of improving the ability of LLMs to detect and correct misleading content in the input itself. In this paper, we propose a novel three-stage fine-tuning method that enhances the ability of LLMs to detect and correct misleading information in the input, further improving response accuracy and reducing hallucinations. Specifically, the three stages include (1) training LLMs to identify misleading information, (2) training LLMs to correct the misleading information using built-in or external knowledge, and (3) training LLMs to generate accurate answers based on the corrected queries. To evaluate our method, we conducted experiments on three datasets for the hallucination detection task and the question answering~(QA) task, as well as two datasets containing misleading information that we constructed. The experimental results demonstrate that our method significantly improves the accuracy and factuality of LLM responses, while also enhancing the ability to detect hallucinations and reducing the generation of hallucinations in the output, particularly when the query contains misleading information.