Yu Zhou, Cheng-Fu Yang, Rui Sun, Di Wu, Sibo Peng, Mingyang Zhang, Zhen Yang, Ruiyi Zhang · 2026 · arXiv
DOI: 10.48550/arXiv.2610.26860 · open-access PDF
多式大语言模型(MLLMs)在视觉推理方面取得了快速的进展,但许多任务需要通过内部推理而难以获得的证据:我们对22个基准点的控制研究表明,仅仅增加基准模型的推断预算只会带来适度的收益。
Machine translation (Meta M2M100); the English original follows.
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 aro
Read in: English · Spanish · Chinese · Hindi · Arabic · Portuguese · French · German · Japanese · Russian · Korean · Indonesian · Vietnamese
GET https://store.agentexchange.work/gw/papers/verify?title=...&year=... (0.003 USD), fetch this record as JSON with /gw/papers/paper?id=2610.26860, or list what it cites with /gw/papers/references?id=.... Pay per call in USDC (x402) or with a 19 USD card key for 1,000 calls. Catalogue · llms.txtSources: OpenAlex (CC0), arXiv, Crossref. Agent Exchange