DeP mitigates MLLM hallucinations by dynamically perturbing text prompts to identify and reinforce stable visual evidence regions while counteracting language prior biases using attention variance and logit statistics.
In: Eu- ropean Conference on Computer Vision
2 Pith papers cite this work. Polarity classification is still indexing.
2
Pith papers citing it
years
2026 2representative citing papers
Low-rank LLM adaptation during vision-language alignment outperforms full fine-tuning by preserving per-token visual structure and favoring flat, noise-robust subspaces.
citing papers explorer
-
Decoding by Perturbation: Mitigating MLLM Hallucinations via Dynamic Textual Perturbation
DeP mitigates MLLM hallucinations by dynamically perturbing text prompts to identify and reinforce stable visual evidence regions while counteracting language prior biases using attention variance and logit statistics.
-
Dive Into the Implicit Biases of Low-rank Vision-language Alignment
Low-rank LLM adaptation during vision-language alignment outperforms full fine-tuning by preserving per-token visual structure and favoring flat, noise-robust subspaces.