Evaluating LLMs' Assessment of Mixed-Context Hallucination Through the Lens of Summarization
Fuente:
arXiv
Salvato in:
| Autori principali: | , , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2025
|
| Soggetti: | |
| Accesso online: | |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866913716459536384 |
|---|---|
| author | Qi, Siya Cao, Rui He, Yulan Yuan, Zheng |
| author_facet | Qi, Siya Cao, Rui He, Yulan Yuan, Zheng |
| contents | With the rapid development of large language models (LLMs), LLM-as-a-judge has emerged as a widely adopted approach for text quality evaluation, including hallucination evaluation. While previous studies have focused exclusively on single-context evaluation (e.g., discourse faithfulness or world factuality), real-world hallucinations typically involve mixed contexts, which remains inadequately evaluated. In this study, we use summarization as a representative task to comprehensively evaluate LLMs' capability in detecting mixed-context hallucinations, specifically distinguishing between factual and non-factual hallucinations. Through extensive experiments across direct generation and retrieval-based models of varying scales, our main observations are: (1) LLMs' intrinsic knowledge introduces inherent biases in hallucination evaluation; (2) These biases particularly impact the detection of factual hallucinations, yielding a significant performance bottleneck; (3) The fundamental challenge lies in effective knowledge utilization, balancing between LLMs' intrinsic knowledge and external context for accurate mixed-context hallucination evaluation. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2503_01670 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Evaluating LLMs' Assessment of Mixed-Context Hallucination Through the Lens of Summarization Qi, Siya Cao, Rui He, Yulan Yuan, Zheng Computation and Language Artificial Intelligence Computers and Society Information Retrieval Machine Learning With the rapid development of large language models (LLMs), LLM-as-a-judge has emerged as a widely adopted approach for text quality evaluation, including hallucination evaluation. While previous studies have focused exclusively on single-context evaluation (e.g., discourse faithfulness or world factuality), real-world hallucinations typically involve mixed contexts, which remains inadequately evaluated. In this study, we use summarization as a representative task to comprehensively evaluate LLMs' capability in detecting mixed-context hallucinations, specifically distinguishing between factual and non-factual hallucinations. Through extensive experiments across direct generation and retrieval-based models of varying scales, our main observations are: (1) LLMs' intrinsic knowledge introduces inherent biases in hallucination evaluation; (2) These biases particularly impact the detection of factual hallucinations, yielding a significant performance bottleneck; (3) The fundamental challenge lies in effective knowledge utilization, balancing between LLMs' intrinsic knowledge and external context for accurate mixed-context hallucination evaluation. |
| title | Evaluating LLMs' Assessment of Mixed-Context Hallucination Through the Lens of Summarization |
| topic | Computation and Language Artificial Intelligence Computers and Society Information Retrieval Machine Learning |
| url | https://arxiv.org/abs/2503.01670 |