MiMIC mitigates visual modality collapse and semantic misalignment in universal multimodal retrieval via fusion-in-decoder architecture and robust single-modality training.
When language overrules: Modality imbalance in vlms
5 Pith papers cite this work. Polarity classification is still indexing.
5
Pith papers citing it
citation-role summary
background 1
citation-polarity summary
fields
cs.CV 5years
2026 5verdicts
UNVERDICTED 5roles
background 1polarities
background 1representative citing papers
MoIR mitigates modality dominance in VLMs by explicitly enriching low-information tokens with routed data from stronger modalities prior to LLM processing, yielding more balanced contributions and improved robustness under degradation.
VLMs fail at counting because visual evidence degrades in later language layers, and a lightweight Modality Attention Share intervention can encourage better use of image information during answer generation.
Filtering post-training data to visually grounded questions improves VLM video understanding performance by up to 6.2 points using 69% of the data.
IPPg embeds text into images to reduce multimodal model inference costs by 35.8-91% with competitive accuracy on many VQA and code benchmarks.
citing papers explorer
-
MiMIC: Mitigating Visual Modality Collapse in Universal Multimodal Retrieval While Avoiding Semantic Misalignment
MiMIC mitigates visual modality collapse and semantic misalignment in universal multimodal retrieval via fusion-in-decoder architecture and robust single-modality training.
-
Information Router for Mitigating Modality Dominance in Vision-Language Models
MoIR mitigates modality dominance in VLMs by explicitly enriching low-information tokens with routed data from stronger modalities prior to LLM processing, yielding more balanced contributions and improved robustness under degradation.
-
Counting to Four is still a Chore for VLMs
VLMs fail at counting because visual evidence degrades in later language layers, and a lightweight Modality Attention Share intervention can encourage better use of image information during answer generation.
-
Watch Before You Answer: Learning from Visually Grounded Post-Training
Filtering post-training data to visually grounded questions improves VLM video understanding performance by up to 6.2 points using 69% of the data.
-
Token-Efficient Multimodal Reasoning via Image Prompt Packaging
IPPg embeds text into images to reduce multimodal model inference costs by 35.8-91% with competitive accuracy on many VQA and code benchmarks.