Search Results for "phase" - Page 2

...Unlike traditional artistic style transfer methods that produce painterly outputs, this approach focuses on maintaining realistic textures, lighting, and spatial consistency. The method is based on a two-step process that includes a stylization phase followed by a smoothing operation, ensuring that the output image remains coherent and free of visual artifacts. It is computationally efficient due to its closed-form solution, allowing fast processing compared to iterative optimization-based methods. The framework is particularly useful in applications such as photo editing, film post-processing, and dataset augmentation where realism is critical. ...

SmartPrediction-ENGINE inside Enhanced with Artificial Intelligence Based upon Game Theory Still This is in BETA Phase

...the expected users are: PHD students and researchers who need to identify relevant bibliographic references in order to carry out literature review as a major and preliminary phase of any research work. Developers are free to adapt it according to their needs, for example in terms of data sources. A knowledge base is used in this project to store aggregated information in OWL, it is available under the name: Citations ontology in the sourceforge platform. These are the emails of developers, do not hesitate to contact them if you have any inquiry about the source code: melissa_am@hotmail.fr chellalinourelhouda@gmail.com lzemmouchi-ghomari@usthb.dz

This is a cross-platform framework for using Genetic Algorithms for solutions. Written in Java and uses convinient plug-in features for every phase in the genetic development, while maintaining an easy-to-use API for easy integration into applications.

...Built on Qwen 2.5-VL-7B and fine-tuned with synthetic Doc → Reasoning → Markdown examples, it generates thinking tokens before producing the final Markdown to better handle complex layouts and tables. It uses a two-phase training process: supervised fine-tuning (SFT) followed by reinforcement learning (GRPO) with a layout-centric reward for accuracy on challenging documents. The model excels at non-standard layouts and complex table structures, outperforming non-reasoning OCR systems like GPT-4o and OCRFlux, and competing with large closed-source reasoning models like Gemini 2.5. ...

Jan-v1-edge is a lightweight agentic language model developed by JanHQ, designed for fast and reliable on-device execution. It is the second release in the Jan Family and was distilled from the larger Jan-v1 model, retaining strong reasoning and problem-solving capabilities while reducing its computational footprint. The model was refined through a two-stage post-training process: Supervised Fine-Tuning (SFT) to transfer knowledge from Jan-v1, followed by Reinforcement Learning with...