腾讯ProLaViT框架使多模态LLM在潜在空间中逐步推理，解决视觉推理失败问题
Tencent ProLaViT Teaches Multimodal LLMs to Reason Step-by-Step in Latent Space, Ending Swallowing-Whole Visual Reasoning Failures
腾讯发了ProLaViT，让多模态模型像人一样一步步想问题，不用外部工具就能解决视觉推理错误，ECCV 2026保底。
腾讯BAC团队提出ProLaViT框架，用于渐进式潜在视觉思维，使多模态LLM（如Qwen2.5-VL）无需外部视觉工具即可在潜在空间中逐步推理。
该框架被ECCV 2026接收，在多个视觉推理基准（如MMMU、MathVista、CLEVR）上相比直接推理方法提升5-15%准确率。ProLaViT通过引入潜在思维令牌，让模型在内部推理步骤中分解复杂视觉问题。
Tencent ProLaViT Teaches Multimodal LLMs to Reason Step-by-Step in Latent Space, Ending Swallowing-Whole Visual Reasoning Failures
Tencent BAC team proposes ProLaViT framework for progressive latent visual thought, enabling multimodal LLMs to reason step-by-step in latent space without external vision tools, accepted at ECCV 2026.
This report is categorized under AITOP's "Models" desk, focusing on China's AI industry and domestic foundation model ecosystem. Within our cross-regional analytical framework, it directly maps to US-China foundation model benchmarks and commercialization comparison. Key entities and technical topics under continuous tracking include ProLaViT, 腾讯, 多模态, 推理模型.
Primary source verification: published directly by Pandaily with first-party authority.
