GEASS selectively gates and weights self-generated captions using confidence and entropy to reduce object hallucinations in VLMs, outperforming vanilla inference and contrastive decoding on POPE and HallusionBench.
arXiv preprint arXiv:2502.03628 , year=
6 Pith papers cite this work. Polarity classification is still indexing.
6
Pith papers citing it
citation-role summary
background 1
citation-polarity summary
roles
background 1polarities
background 1representative citing papers
Prefill-Time Intervention (PTI) reduces hallucinations in large vision-language models by applying a one-time modality-aware steering correction to the initial KV cache at the prefill stage rather than during autoregressive decoding.
Decoder-based VLMs hallucinate due to geometric over-alignment of visual embeddings with the text manifold in a universal dataset-agnostic subspace, mitigated by projecting out the linguistic bias.
ACE uses adversarial counter-commonsense perturbations on image tokens during decoding to suppress hallucinated linguistic priors while preserving stable visual signals in MLLMs.
The survey organizes causes of hallucinations in MLLMs, reviews evaluation benchmarks and metrics, and outlines mitigation approaches plus open questions.
Steering is positioned as a distinct adaptation paradigm that uses targeted activation interventions for local, reversible behavioral changes without parameter updates.
citing papers explorer
-
GEASS: Training-Free Caption Steering for Hallucination Mitigation in Vision-Language Models
GEASS selectively gates and weights self-generated captions using confidence and entropy to reduce object hallucinations in VLMs, outperforming vanilla inference and contrastive decoding on POPE and HallusionBench.
-
Prefill-Time Intervention for Mitigating Hallucination in Large Vision-Language Models
Prefill-Time Intervention (PTI) reduces hallucinations in large vision-language models by applying a one-time modality-aware steering correction to the initial KV cache at the prefill stage rather than during autoregressive decoding.
-
When Language Overwrites Vision: Over-Alignment and Geometric Debiasing in Vision-Language Models
Decoder-based VLMs hallucinate due to geometric over-alignment of visual embeddings with the text manifold in a universal dataset-agnostic subspace, mitigated by projecting out the linguistic bias.
-
Not Blind but Silenced: Rebalancing Vision and Language via Adversarial Counter-Commonsense Equilibrium
ACE uses adversarial counter-commonsense perturbations on image tokens during decoding to suppress hallucinated linguistic priors while preserving stable visual signals in MLLMs.
-
From Weights to Activations: Is Steering the Next Frontier of Adaptation?
Steering is positioned as a distinct adaptation paradigm that uses targeted activation interventions for local, reversible behavioral changes without parameter updates.