Ensuring the safety of generative MLLMs is absolutely crucial in order to prevent harm, build trust, address ethical concerns, and enable their responsible deployment in real-world applications. Our results demonstrate that Granite Vision performs almost at par with baselines (despite being the lightest MLLM in the comparison pool) for VLM-as-a-Judge task. Notably, the addition of Safety Vectors to Granite Vision leads to a significant improvement in safety classification performance. We do acknowledge that further work needs to be done to improve high-level reasoning and correct occasional incorrect outputs to improve reliability in sensitive tasks, which require nuanced classification. To address these, we will incorporate more reasoning-focused and structure-related data into the training process in the future. In addition, we showed in this paper that finding safety vectors (SVs) in Granite Vision's attention heads led to significant improvements when safety tasks were reformulated as classification problems. Current reliance for SVs is on few-shot samples which are informative but may have limited scope in terms of capturing the range of possible safety issues that can be encountered. To further improve the model's ability to identify and address all safety concerns, we plan to investigate scaling up SVs using more training data in future research.

Deep Neural Networks (DNNs) have been widely used for various applications, such as image classification [22; 40], object segmentation [33; 35], and object detection [6; 43]. However, the increasing size and complexity of DNNs often result in substantial computational and memory requirements, posing challenges for deployment on resource-constrained platforms, such as mobile or embedded devices. Consequently, developing efficient methods to reduce the computational complexity and storage demands of large models, while minimizing performance degradation, has become essential. Network pruning is one of the most popular methods in model compression. Specifically, current network pruning methods are categorized into unstructured and structured pruning [5]. Unstructured pruning [11; 24] focuses on eliminating individual weights from a network to create fine-grained sparsity. Although these approaches achieve an excellent balance between model size reduction and accuracy retention, they often require specific hardware support for acceleration, which is impractical for general-purpose computing environments. Conversely, structured pruning [23; 18; 29] avoids these hardware dependencies by eliminating redundant network structures, thus introducing a more manageable and hardware-compatible form of sparsity. As a result, structured pruning has become popular and is extensively utilized.

Large Language Models (LLMs) trained on code are revolutionizing the software development process. Increasingly, code LLMs are being integrated into software development environments to improve the productivity of human programmers, and LLM-based agents are beginning to show promise for handling complex tasks autonomously. Realizing the full potential of code LLMs requires a wide range of capabilities, including code generation, fixing bugs, explaining and documenting code, maintaining repositories, and more. In this work, we introduce the Granite series of decoder-only code models for code generative tasks, trained with code written in 116 programming languages. The Granite Code models family consists of models ranging in size from 3 to 34 billion parameters, suitable for applications ranging from complex application modernization tasks to on-device memory-constrained use cases. Evaluation on a comprehensive set of tasks demonstrates that Granite Code models consistently reaches state-of-the-art performance among available open-source code LLMs. The Granite Code model family was optimized for enterprise software development workflows and performs well across a range of coding tasks (e.g. code generation, fixing and explanation), making it a versatile all around code model. We release all our Granite Code models under an Apache 2.0 license for both research and commercial use.