Parenting: Optimizing knowledge selection of retrieval-augmented language models with parameter decoupling and tailored tuning
Retrieval-Augmented Generation (RAG) offers an effective solution to the issues faced by
Large Language Models (LLMs) in hallucination generation and knowledge obsolescence …
Large Language Models (LLMs) in hallucination generation and knowledge obsolescence …
Mitigating copy bias in in-context learning through neuron pruning
Large language models (LLMs) have demonstrated impressive few-shot in-context learning
(ICL) abilities. Still, we show that they are sometimes prone to acopying bias', where they …
(ICL) abilities. Still, we show that they are sometimes prone to acopying bias', where they …
Towards understanding multi-task learning (generalization) of llms via detecting and exploring task-specific neurons
While large language models (LLMs) have demonstrated superior multi-task capabilities,
understanding the learning mechanisms behind this is still a challenging problem. In this …
understanding the learning mechanisms behind this is still a challenging problem. In this …
Dynamic Attention-Guided Context Decoding for Mitigating Context Faithfulness Hallucinations in Large Language Models
Y Huang, Y Zhang, N Cheng, Z Li, S Wang… - arxiv preprint arxiv …, 2025 - arxiv.org
Large language models (LLMs) often suffer from context faithfulness hallucinations, where
outputs deviate from retrieved information due to insufficient context utilization and high …
outputs deviate from retrieved information due to insufficient context utilization and high …
The Scaling Law for LoRA Base on Mutual Information Upper Bound
J Zhang, H Gao, P Zhang, S Sun, C Yang… - arxiv preprint arxiv …, 2025 - arxiv.org
LoRA (Low-Rank Adaptation) is a widely used model fine-tuning method. In fine-tuning, the
law among model performance, model parameters, and data complexity has been a focal …
law among model performance, model parameters, and data complexity has been a focal …