Introduction Understanding how the brain builds internal representations of the visual world is one of the most fascinating challenges in neuroscience. Over the past decade, deep learning has reshaped computer vision, producing neural networks that not only perform at human-level accuracy on recognition tasks but also seem to process information in ways that resemble our…

Introduction Vision Language Models (VLMs) allow both text inputs and visual understanding. However, image resolution is crucial for VLM performance for processing text and chart-rich data. Increasing image resolution creates significant challenges. First, pretrained vision encoders often struggle with high-resolution images due to inefficient pretraining requirements. Running inference on high-resolution images increases computational costs and…

In the domain of multimodal AI, instruction-based image editing models are transforming how users interact with visual content. Just released in August 2025 by Alibaba’s Qwen Team, Qwen-Image-Edit builds on the 20B-parameter Qwen-Image foundation to deliver advanced editing capabilities. This model excels in semantic editing (e.g., style transfer and novel view synthesis) and appearance editing…

Contrastive Language-Image Pre-training (CLIP) has become important for modern vision and multimodal models, enabling applications such as zero-shot image classification and serving as vision encoders in MLLMs. However, most CLIP variants, including Meta CLIP, are limited to English-only data curation, ignoring a significant amount of non-English content from the worldwide web. Scaling CLIP to include…

Multimodal reasoning, where models integrate and interpret information from multiple sources such as text, images, and diagrams, is a frontier challenge in AI. VL-Cogito is a state-of-the-art Multimodal Large Language Model (MLLM) proposed by DAMO Academy (Alibaba Group) and partners, introducing a robust reinforcement learning pipeline that fundamentally upgrades the reasoning skills of large models…

Estimated reading time: 5 minutes Introduction Embodied AI agents are increasingly being called upon to interpret complex, multimodal instructions and act robustly in dynamic environments. ThinkAct, presented by researchers from Nvidia and National Taiwan University, offers a breakthrough for vision-language-action (VLA) reasoning, introducing reinforced visual latent planning to…

Vision Language Models (VLMs) allow both text inputs and visual understanding. However, image resolution is crucial for VLM performance for processing text and chart-rich data. Increasing image resolution creates significant challenges. First, pretrained vision encoders often struggle with high-resolution images due to inefficient pretraining requirements. Running inference on high-resolution images increases computational costs and latency…

Multimodal foundation models (MFMs) like GPT-4o, Gemini, and Claude have shown rapid progress recently, especially in public demos. While their language skills are well studied, their true ability to understand visual information remains unclear. Most benchmarks used today focus heavily on text-based tasks, such as VQA or classification, which often reflect language strengths more than…

Navigating the dense urban canyons of cities like San Francisco or New York can be a nightmare for GPS systems. The towering skyscrapers block and reflect satellite signals, leading to location errors of tens of meters. For you and me, that might mean a missed turn. But for an autonomous vehicle or a delivery robot,…

Process-supervised reward models (PRMs) offer fine-grained, step-wise feedback on model responses, aiding in selecting effective reasoning paths for complex tasks. Unlike output reward models (ORMs), which evaluate responses based on final outputs, PRMs provide detailed assessments at each step, making them particularly valuable for reasoning-intensive applications. While PRMs have been extensively studied in language tasks,…