金·卡戴珊称ChatGPT“亦敌亦友”
"幻觉" 相关文章
LLM Hallucinations: An Internal Tug of War
Teaching LLMs to Abstain via Fine-Grained Semantic Confidence Reward
Neural Diversity Regularizes Hallucinations in Small Models
JT-Safe: Intrinsically Enhancing the Safety and Trustworthiness of LLMs
Knowledge Reasoning Language Model: Unifying Knowledge and Language for Inductive Knowledge Graph Reasoning
Multi-stage Prompt Refinement for Mitigating Hallucinations in Large Language Models
Trustworthy Retrosynthesis: Eliminating Hallucinations with a Diverse Ensemble of Reaction Scorers
Evaluating Hallucinations in Multimodal LLMs with Spoken Queries under Diverse Acoustic Conditions
Evaluating Hallucinations in Multimodal LLMs with Spoken Queries under Diverse Acoustic Conditions
Evaluating Hallucinations in Multimodal LLMs with Spoken Queries under Diverse Acoustic Conditions
EMNLP 2025 | 拨云见日：知识电路分析揭示大语言模型“知识遮蔽”幻觉之源
EMNLP 2025 | 拨云见日：知识电路分析揭示大语言模型“知识遮蔽”幻觉之源
EMNLP 2025 | 拨云见日：知识电路分析揭示大语言模型“知识遮蔽”幻觉之源
ChainMPQ: Interleaved Text-Image Reasoning Chains for Mitigating Relation Hallucinations
Distributional Semantics Tracing: A Framework for Explaining Hallucinations in Large Language Models
Hallucination is Inevitable for LLMs with the Open World Assumption
因报告中发现AI幻觉 德勤同意向澳大利亚政府退款
SECA: Semantically Equivalent and Coherent Attacks for Eliciting LLM Hallucinations
Culture, Language, and Generative Language Models
