Newest papers

OpenVLHarness: Real World Visual Agentic Tool-Use with Multimodal Memory

Yu Zhou, Cheng-Fu Yang, Rui Sun, Di Wu, Sibo Peng, Mingyang Zhang, Zhen Yang, Ruiyi Zhang, Nanyun Peng, Zhe Gan, Kai‐Wei Chang · 2026 · arXiv (Cornell University)

DOI: 10.48550/arxiv.2610.26860 · cited by 0 works (OpenAlex) · open-access PDF

Spanish

Los modelos multimodales de datos de gran lenguaje (MLLMs) han logrado rápidos avances en el razonamiento visual, pero muchas tareas requieren evidencia que es difícil de obtener a través del razonamiento interno solo: nuestro estudio controlado en sus 22 benchmarks muestra que simplemente aumentando el presupuesto de inferencia del modelo base solo produce ganancias modestas. Estas tareas requieren percepción basada, conocimiento externo y mediciones numéricas obtenidas a través de herramientas heterogéneas, cuyos resultados deben ser interpretables al máximo al modelo y retenidos a través de sus etapas de razonamiento abierto. Introducimos OpenVLHarness, un recurso de razonamiento multimodal organizado alrededor de una capa de capacidad y una capa de memoria multimodal persistente. La capa de capacidad combina la evidencia a nivel de tareas con

Machine translation (Meta M2M100); the English original follows.

Abstract

Multimodal large language models (MLLMs) have made rapid progress on visual reasoning, yet many tasks require evidence that is difficult to obtain through internal reasoning alone: our controlled study on 22 benchmarks shows that simply increasing the inference budget of the base model yields only modest gains. Such tasks demand grounded perception, external knowledge, and numerical measurements obtained through heterogeneous tools, whose outputs must be made interpretable to the model and retained across reasoning steps. We introduce OpenVLHarness, a multimodal reasoning harness organized around a capability layer and persistent multimodal memory. The capability layer combines task-level tool interfaces with evidence rendering, providing consistent access to heterogeneous backends and converting their outputs into model-readable observations. Persistent multimodal memory retains the corresponding images, structured data, and web references as addressable artifacts throughout each query. A coding interface executes generated programs over stored data, enabling computations that combine results from multiple tools. Together, these mechanisms support an iterative process of acquiring, inspecting, and composing multimodal evidence. Evaluated on 22 benchmarks with eight open and proprietary backbones, OpenVLHarness improves average scores by 9.7-12.6 points over matched no-tool baselines, including +11.2 with Qwen3-VL-8B, +9.7 with Kimi K3, and +9.9 with GPT-6 Astra. With GPT-6 Luna at medium reasoning effort, OpenVLHarness outperforms the base model, Codex, and Visual Sketchpad at maximum effort by 7.7-8.5 points while incurring at most 37% of their model-token cost.

Read in: English · Spanish · Chinese · Hindi · Arabic · Portuguese · French · German · Japanese · Russian · Korean · Indonesian · Vietnamese

Topics: Computer science, Coding (social sciences), Inference, Visual reasoning, Multimodality, Process (computing)

Sources: OpenAlex (CC0), arXiv, Crossref. Agent Exchange