Look Closer! An Adversarial Parametric Editing Framework for Hallucination Mitigation in VLMs (AAAI 2025) 2026-02-14 #深度学习 #大模型
CRoPS: A Training-Free Hallucination Mitigation Framework for Vision-Language Models (TMLR 2026) 2026-02-14 #深度学习 #大模型
REVIS:Sparse Latent Steering to Mitigate Object Hallucination in Large Vision-Language Models (arxiv 2026) 和VTI差不多。 2026-02-13 #深度学习 #大模型
microgpt AI 领域大神 Andrej Karpathy 编写。展示了如何仅使用 Python 标准库(无 PyTorch、NumPy、TensorFlow),从零构建并训练一个完整的 GPT模型。 2026-02-12 #大模型
Light Alignment Improves LLM Safety via Model Self-Reflection with a Single Neuron (arxiv 2026) 2026-02-12 #深度学习 #大模型
FraudShield: Knowledge Graph Empowered Defense for LLMs against Fraud Attacks (WWW 2026) 2026-02-11 #深度学习 #大模型
PoisonedEye:Knowledge Poisoning Attack on Retrieval-Augmented Generation based Large Vision-Language Models (ICML 2025) 2026-02-09 #深度学习 #RAG
TraceDet: Hallucination Detection from the Decoding Trace of Diffusion Large Language Models (ICLR 2026) 2026-02-09 #深度学习 #大模型