# OPDAI at SemEval-2024 Task 6: Small LLMs can Accelerate Hallucination Detection with Weakly Supervised Data

arXiv: https://arxiv.org/abs/2402.12913 · HTML: https://mlsanity.com/paper/2402.12913/

## Key facts

- Cited by 3 papers in the map (data through 14 Oct 2026).
- Builds on 7 of its 25 references that are in the map.
- First posted on arXiv 20 Feb 2024.

## Builds on (most cited first, 7 of 7)

- [LoRA: Low-Rank Adaptation of Large Language Models](https://mlsanity.com/paper/2106.09685/) (2021)
- [Self-Consistency Improves Chain of Thought Reasoning in Language Models](https://mlsanity.com/paper/2203.11171/) (2022)
- [Model soups: averaging weights of multiple fine-tuned models improves accuracy without increasing inference time](https://mlsanity.com/paper/2203.05482/) (2022)
- [Risk Taxonomy, Mitigation, and Assessment Benchmarks of Large Language Model Systems](https://mlsanity.com/paper/2401.05778/) (2024)
- [A Survey on Ensemble Learning under the Era of Deep Learning](https://mlsanity.com/paper/2101.08387/) (2021)
- [mLoRA: Fine-Tuning LoRA Adapters via Highly-Efficient Pipeline Parallelism in Multiple GPUs](https://mlsanity.com/paper/2312.02515/) (2023)
- [Using Deep Mixture-of-Experts to Detect Word Meaning Shift for TempoWiC](https://mlsanity.com/paper/2211.03466/) (2022)

## Built on by (most cited first, 3 of 3)

- [Arcee's MergeKit: A Toolkit for Merging Large Language Models](https://mlsanity.com/paper/2403.13257/) (2024)
- [SemEval-2024 Shared Task 6: SHROOM, a Shared-task on Hallucinations and Related Observable Overgeneration Mistakes](https://mlsanity.com/paper/2403.07726/) (2024)
- [HalluSearch at SemEval-2025 Task 3: A Search-Enhanced RAG Pipeline for Hallucination Detection](https://mlsanity.com/paper/2504.10168/) (2025)

## Abstract (arXiv)

This paper mainly describes a unified system for hallucination detection of LLMs, which wins the second prize in the model-agnostic track of the SemEval-2024 Task 6, and also achieves considerable results in the model-aware track. This task aims to detect hallucination with LLMs for three different text-generation tasks without labeled training data. We utilize prompt engineering and few-shot learning to verify the performance of different LLMs on the validation data. Then we select the LLMs with better performance to generate high-quality weakly supervised training data, which not only satisfies the consistency of different LLMs, but also satisfies the consistency of the optimal LLM with different sampling parameters. Furthermore, we finetune different LLMs by using the constructed training data, and finding that a relatively small LLM can achieve a competitive level of performance in hallucination detection, when compared to the large LLMs and the prompt-based approaches using GPT-4.