KidVis: Do Multimodal Large Language Models Possess the Visual Perceptual Capabilities of a 6-Year-Old?

OpenAI has launched GPT-5.2, its latest AI model, which is designed to enhance productivity and has shown mixed results in tests compared to its predecessor, GPT-5.1. This development comes amid increasing competition from Google's Gemini 3, which has rapidly gained a significant user base.

The introduction of Time Puzzles marks a significant advancement in evaluating iterative temporal reasoning in large language models (LLMs). This task combines factual temporal anchors with cross-cultural calendar relations, generating puzzles that challenge LLMs' reasoning capabilities. Despite the simplicity of the dataset, models like GPT-5 achieved only 49.3% accuracy, highlighting the difficulty of the task.

A recent study has explored the integration of visual and textual information in Multimodal Large Language Models (MLLMs), revealing that visual-text fusion occurs at specific layers within these models rather than uniformly across the network. The research highlights a late-stage

A recent study has introduced a framework aimed at mitigating hallucination issues in Multimodal Large Language Models (MLLMs) during Reinforcement Learning (RL) optimization. The research identifies key factors contributing to hallucinations, including over-reliance on visual reasoning and insufficient exploration diversity. The proposed framework incorporates modules for caption feedback, diversity-aware sampling, and conflict regularization to enhance model reliability.

A new framework called From Rows to Reasoning (FRTR) has been introduced to enhance the reasoning capabilities of Large Language Models (LLMs) when dealing with complex spreadsheets. This framework includes FRTR-Bench, a benchmark featuring 30 enterprise-grade Excel workbooks, which aims to improve the understanding of multimodal data by breaking down spreadsheets into granular components.

A new method called PRISM has been introduced to optimize the selection of training data for Multimodal Large Language Models (MLLMs), addressing the redundancy in rapidly growing datasets that increases computational costs. This self-pruning intrinsic selection method aims to enhance efficiency without the need for extensive training or proxy-based inference techniques.

A recent study introduced MoHoBench, a benchmark designed to assess the honesty of Multimodal Large Language Models (MLLMs) when confronted with unanswerable visual questions. This research highlights the need for a systematic evaluation of MLLMs' response behaviors, as their trustworthiness in generating content remains underexplored.