Hallucinate Less by Thinking More: Aspect-Based Causal Abstention for Large Language Models
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_ | 1866915630799650816 |
|---|---|
| author | Nguyen, Vy Xu, Ziqi Chan, Jeffrey He, Estrid Xia, Feng Zhang, Xiuzhen |
| author_facet | Nguyen, Vy Xu, Ziqi Chan, Jeffrey He, Estrid Xia, Feng Zhang, Xiuzhen |
| contents | Large Language Models (LLMs) often produce fluent but factually incorrect responses, a phenomenon known as hallucination. Abstention, where the model chooses not to answer and instead outputs phrases such as "I don't know", is a common safeguard. However, existing abstention methods typically rely on post-generation signals, such as generation variations or feedback, which limits their ability to prevent unreliable responses in advance. In this paper, we introduce Aspect-Based Causal Abstention (ABCA), a new framework that enables early abstention by analysing the internal diversity of LLM knowledge through causal inference. This diversity reflects the multifaceted nature of parametric knowledge acquired from various sources, representing diverse aspects such as disciplines, legal contexts, or temporal frames. ABCA estimates causal effects conditioned on these aspects to assess the reliability of knowledge relevant to a given query. Based on these estimates, we enable two types of abstention: Type-1, where aspect effects are inconsistent (knowledge conflict), and Type-2, where aspect effects consistently support abstention (knowledge insufficiency). Experiments on standard benchmarks demonstrate that ABCA improves abstention reliability, achieves state-of-the-art performance, and enhances the interpretability of abstention decisions. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2511_17170 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Hallucinate Less by Thinking More: Aspect-Based Causal Abstention for Large Language Models Nguyen, Vy Xu, Ziqi Chan, Jeffrey He, Estrid Xia, Feng Zhang, Xiuzhen Computation and Language Artificial Intelligence Large Language Models (LLMs) often produce fluent but factually incorrect responses, a phenomenon known as hallucination. Abstention, where the model chooses not to answer and instead outputs phrases such as "I don't know", is a common safeguard. However, existing abstention methods typically rely on post-generation signals, such as generation variations or feedback, which limits their ability to prevent unreliable responses in advance. In this paper, we introduce Aspect-Based Causal Abstention (ABCA), a new framework that enables early abstention by analysing the internal diversity of LLM knowledge through causal inference. This diversity reflects the multifaceted nature of parametric knowledge acquired from various sources, representing diverse aspects such as disciplines, legal contexts, or temporal frames. ABCA estimates causal effects conditioned on these aspects to assess the reliability of knowledge relevant to a given query. Based on these estimates, we enable two types of abstention: Type-1, where aspect effects are inconsistent (knowledge conflict), and Type-2, where aspect effects consistently support abstention (knowledge insufficiency). Experiments on standard benchmarks demonstrate that ABCA improves abstention reliability, achieves state-of-the-art performance, and enhances the interpretability of abstention decisions. |
| title | Hallucinate Less by Thinking More: Aspect-Based Causal Abstention for Large Language Models |
| topic | Computation and Language Artificial Intelligence |
| url | https://arxiv.org/abs/2511.17170 |