LVLM-Intrepret: an interpretability tool for large vision-language models

G Ben Melech Stan, E Aflalo… - Proceedings of the …, 2024 - openaccess.thecvf.com
In the rapidly evolving landscape of artificial intelligence multi-modal large language models
are emerging as a significant area of interest. These models which combine various forms of …

LVLM-Intrepret: An Interpretability Tool for Large Vision-Language Models

GBM Stan, RY Rohekar, Y Gurwicz, ML Olson… - arXiv preprint arXiv …, 2024 - arxiv.org
In the rapidly evolving landscape of artificial intelligence, multi-modal large language
models are emerging as a significant area of interest. These models, which combine various …

DeCo: Decoupling Token Compression from Semantic Abstraction in Multimodal Large Language Models

L Yao, L Li, S Ren, L Wang, Y Liu, X Sun… - arXiv preprint arXiv …, 2024 - arxiv.org
The visual projector, which bridges the vision and language modalities and facilitates cross-
modal alignment, serves as a crucial component in MLLMs. However, measuring the …

CopyLens: Dynamically Flagging Copyrighted Sub-Dataset Contributions to LLM Outputs

Q Ma, RJ Zhu, P Liu, R Yan, F Zhang, L Liang… - arXiv preprint arXiv …, 2024 - arxiv.org
Large Language Models (LLMs) have become pervasive due to their knowledge absorption
and text-generation capabilities. Concurrently, the copyright issue for pretraining datasets …

Probabilistic Conceptual Explainers: Trustworthy Conceptual Explanations for Vision Foundation Models

H Wang, S Tan, H Wang - arXiv preprint arXiv:2406.12649, 2024 - arxiv.org
Vision transformers (ViTs) have emerged as a significant area of focus, particularly for their
capacity to be jointly trained with large language models and to serve as robust vision …

Deep Causal Generative Models with Property Control

Q Zhao, S Wang, G Bai, B Pan, Z Qin… - arXiv preprint arXiv …, 2024 - arxiv.org
Generating data with properties of interest by external users while following the right
causation among its intrinsic factors is important yet has not been well addressed jointly …

Logic Synthesis Optimization with Predictive Self-Supervision via Causal Transformers

R Karimi, F Faez, Y Zhang, X Li, L Chen… - arXiv preprint arXiv …, 2024 - arxiv.org
Contemporary hardware design benefits from the abstraction provided by high-level logic
gates, streamlining the implementation of logic circuits. Logic Synthesis Optimization (LSO) …

[PDF][PDF] Exploring Explainable NLP Techniques for Trait Extraction and Personality Inference

OA Saad, M Abuelkheir - 2024 - researchgate.net
Explaining AI models' predictions is a crucial step to ensure their trustworthiness and help
users understand a model's behavior. In this thesis, we focus on explaining the predictions …