FaithLens: Detecting and Explaining Faithfulness Hallucination
Fuente:
arXiv
Enregistré dans:
| Auteurs principaux: | , , , , , , , , , , |
|---|---|
| Format: | Preprint |
| Publié: |
2025
|
| Sujets: | |
| Accès en ligne: | |
| Tags: |
Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
|
| _version_ | 1866918458409615360 |
|---|---|
| author | Si, Shuzheng Wang, Qingyi Zhao, Haozhe Bai, Yuzhuo Chen, Guanqiao Luo, Kangyang Chen, Gang Qi, Fanchao Zhang, Minjia Chang, Baobao Sun, Maosong |
| author_facet | Si, Shuzheng Wang, Qingyi Zhao, Haozhe Bai, Yuzhuo Chen, Guanqiao Luo, Kangyang Chen, Gang Qi, Fanchao Zhang, Minjia Chang, Baobao Sun, Maosong |
| contents | Recognizing whether outputs from large language models (LLMs) contain faithfulness hallucination is crucial for real-world applications, e.g., retrieval-augmented generation and summarization. In this paper, we introduce FaithLens, a cost-efficient and effective faithfulness hallucination detection model that can jointly provide binary predictions and corresponding explanations to improve trustworthiness. To achieve this, we first synthesize training data with explanations via advanced LLMs and apply a well-defined data filtering strategy to ensure label correctness, explanation quality, and data diversity. Subsequently, we fine-tune the model on these well-curated training data as a cold start and further optimize it with rule-based reinforcement learning, using rewards for both prediction correctness and explanation quality. Results on 12 diverse tasks show that the 8B-parameter FaithLens outperforms advanced models such as GPT-5.2 and o3. Also, FaithLens can produce high-quality explanations, delivering a distinctive balance of trustworthiness, efficiency, and effectiveness. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2512_20182 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | FaithLens: Detecting and Explaining Faithfulness Hallucination Si, Shuzheng Wang, Qingyi Zhao, Haozhe Bai, Yuzhuo Chen, Guanqiao Luo, Kangyang Chen, Gang Qi, Fanchao Zhang, Minjia Chang, Baobao Sun, Maosong Computation and Language Artificial Intelligence Recognizing whether outputs from large language models (LLMs) contain faithfulness hallucination is crucial for real-world applications, e.g., retrieval-augmented generation and summarization. In this paper, we introduce FaithLens, a cost-efficient and effective faithfulness hallucination detection model that can jointly provide binary predictions and corresponding explanations to improve trustworthiness. To achieve this, we first synthesize training data with explanations via advanced LLMs and apply a well-defined data filtering strategy to ensure label correctness, explanation quality, and data diversity. Subsequently, we fine-tune the model on these well-curated training data as a cold start and further optimize it with rule-based reinforcement learning, using rewards for both prediction correctness and explanation quality. Results on 12 diverse tasks show that the 8B-parameter FaithLens outperforms advanced models such as GPT-5.2 and o3. Also, FaithLens can produce high-quality explanations, delivering a distinctive balance of trustworthiness, efficiency, and effectiveness. |
| title | FaithLens: Detecting and Explaining Faithfulness Hallucination |
| topic | Computation and Language Artificial Intelligence |
| url | https://arxiv.org/abs/2512.20182 |