Knowledge Tree
Relations (1)
related 3.17 — strongly supporting 8 facts
Hallucinations are defined as a core behavioral issue within language models [1], serving as a significant obstacle to their reliability [2]. Research consistently explores how language model design choices, such as decoding methods [3], temperature settings [4], and training paradigms {fact:6, fact:7, fact:8}, directly influence the occurrence of these false statements.
Facts (8)
Sources
Evaluating Evaluation Metrics -- The Mirage of Hallucination Detection arxiv.org 3 facts
claimMode-seeking decoding methods appear to reduce hallucinations in language models, particularly in knowledge-grounded settings.
claimHallucinations are a significant obstacle to the reliability and widespread adoption of language models.
claimThe accurate measurement of hallucinations remains a persistent challenge for language models despite the proposal of many task- and domain-specific metrics.
Awesome-Hallucination-Detection-and-Mitigation - GitHub github.com 2 facts
referenceThe paper 'Unfamiliar Finetuning Examples Control How Language Models Hallucinate' by Kang et al. (2024) investigates the impact of finetuning examples on hallucination behavior.
referenceThe paper "Do I Know This Entity? Knowledge Awareness and Hallucinations in Language Models" by Ferrando et al. (2025) investigates the relationship between knowledge awareness and the occurrence of hallucinations in language models.
Why language models hallucinate | OpenAI openai.com 1 fact
claimHallucinations in language models are defined as plausible but false statements generated by the models.
[2509.04664] Why Language Models Hallucinate - arXiv arxiv.org 1 fact
claimLanguage models persist in hallucinating because they are optimized to be good test-takers, and guessing when uncertain improves performance on most current evaluation benchmarks.
Empowering RAG Using Knowledge Graphs: KG+RAG = G-RAG neurons-lab.com 1 fact
claimSetting language model temperature parameters to zero reduces the likelihood of hallucination, but it is insufficient to eliminate the issue because language models are inherently designed to predict the next token.