Challenge: Vision-Language Models (VLMs) provide a unified framework to process both text-only and vision-language tasks.
Approach: They propose a method to reduce the distance between visual and textual representations by introducing a Representation Distribution Difference (RDD) loss.
Outcome: Empirical evidence shows that finetuning VLMs on vision-language data has degraded language capabilities.

Similar Papers

Unraveling and Mitigating Safety Alignment Degradation of Vision-Language Models (2025.findings-acl)

Copied to clipboard

Challenge: LLaVA-7B demonstrated a decline in safety alignment ability on multi-modal inputs compared to its LLM backbone.
Approach: They propose a method to recover alignment ability from LLM backbone while preserving functional capabilities of VLMs.
Outcome: The proposed framework recovers alignment ability that is inherent in the LLM backbone with minimal impact on fluency and linguistic capabilities of pre-trained VLMs.
Lost in Embeddings: Information Loss in Vision–Language Models (2025.findings-emnlp)

Copied to clipboard

Challenge: Experiments reveal connectors substantially distort the local geometry of visual representations, with k-nearest neighbors diverging by 40–60% post-projection, correlating with degradation in retrieval performance.
Approach: They propose two approaches to examine and quantify information loss by analyzing latent representation space.
Outcome: The proposed model improves retrieval performance by analyzing changes in k-nearest neighbor relationships between image representations before and after projection.
LLMs Can Compensate for Deficiencies in Visual Representations (2025.findings-emnlp)

Copied to clipboard

Challenge: a strong language backbone in vision-language models compensates for weak visual features by contextualizing or enriching them.
Approach: They investigate whether strong language backbone compensates for weak visual features . they use CLIP-based vision encoders to perform controlled self-attention ablations .
Outcome: The proposed model compensates for weak visual features by contextualizing or enriching them.
RWKV-CLIP: A Robust Vision-Language Representation Learner (2024.emnlp-main)

Copied to clipboard

Challenge: Using large image-text datasets, large-scale image-data sets have been used for visionlanguage pre-training.
Approach: They propose a framework that leverages Large Language Models to combine and refine information from web-based image-text pairs, synthetic captions, and detection tags.
Outcome: The proposed framework can combine and refine information from web-based image-text pairs, synthetic captions, and detection tags.
Expedited Training of Visual Conditioned Language Generation via Redundancy Reduction (2024.acl-long)

Copied to clipboard

Challenge: EVLGen is a framework for visual-language pre-training with high computational demands.
Approach: They propose a streamlined framework for the pre-training of visually conditioned language generation models with high computational demands.
Outcome: The proposed framework accelerates training of vision-language models by a factor of 5 without compromising performance.
Red Teaming Visual Language Models (2024.findings-acl)

Copied to clipboard

Challenge: VLMs (Vision-Language Models) can be induced to generate harmful or inaccurate content through specific test cases.
Approach: They propose a red teaming dataset which encompasses 12 subtasks under 4 primary aspects (faithfulness, privacy, safety, fairness) this dataset is the first to benchmark current VLMs in terms of these 4 aspects .
Outcome: The proposed dataset shows that 10 open-source VLMs struggle with red teaming in different degrees and have up to 31% performance gap with GPT-4V.
Vision-Language Models Align with Human Neural Representations in Concept Processing (2026.eacl-long)

Copied to clipboard

Challenge: Recent studies suggest that transformer-based vision-language models capture the multimodality of concept processing in the human brain.
Approach: They analysed multiple VLMs employing different strategies to integrate visual and textual modalities, along with language-only counterparts.
Outcome: The transformer-based vision-language models outperform language-only models in two experimental conditions, while only some outperformed the language-based models.
Can VLMs Actually See and Read? A Survey on Modality Collapse in Vision-Language Models (2025.findings-acl)

Copied to clipboard

Challenge: Vision-language models integrate textual and visual information, enabling them to process visual inputs and generate predictions.
Approach: They review work on modality collapse analysis to provide insights into the reason for this unintended behavior and review probing studies for fine-grained vision-language understanding.
Outcome: The proposed models can achieve competitive performance in vision-language tasks despite relying heavily on textual information and ignoring visual information.
Seeing Through Words, Speaking Through Pixels: Deep Representational Alignment Between Vision and Language Models (2025.emnlp-main)

Copied to clipboard

Challenge: Recent studies show that deep vision-only and language-only models project inputs into a partially aligned representational space.
Approach: They investigate whether a model's representational code is semantically shared . they find that alignment peaks in mid-to-late layers of both model types .
Outcome: a forced-choice "Pick-a-Pic" task shows human preferences for image-caption matches are mirrored in embedding spaces across vision-language model pairs.
Measuring Progress in Fine-grained Vision-and-Language Understanding (2023.acl-long)

Copied to clipboard

Challenge: X-VLM models lack "fine-grained" understanding of relationships, verbs and numbers in images . pretraining on large-scale image–text data from the Web has facilitated rapid progress on many vision-and-language tasks .
Approach: They investigate models that outperform other baselines on fine-grained data . they highlight importance of novel losses and rich data sources for learning fine-grain skills .
Outcome: The proposed model outperforms baseline models on four fine-grained benchmarks . the model outpersforms other baseline models and even degrades performance .

What is GenGO?

GenGO is an NLP powered publication search system. It currenctly indexes 30k+ papers from ACL Anthology, and implements multi-aspect summarization, semantic search, and more!

Information

About
Limitations