Abstract
. Current multimodal latent reasoning often relies on external supervision (e.g., auxiliary images), ignoring intrinsic visual attention dynamics. In this work, we identify a critical Perception Gap in distillation: student models frequently mimic a teacher's textual output while attending to fundamentally divergent visual regions, effectively relying on language priors rather than grounded perception. To bridge this, we propose LaViT, a framework that aligns latent visual thoughts rather than static embeddings. LaViT compels the student to autoregressively reconstruct the teacher's visual semantics and attention trajectories prior to text generation, employing a curriculum sensory gating mechanism to prevent shortcut learning. Extensive experiments show that LaViT significantly enhances visual grounding, achieving up to +16.9% gains on complex reasoning tasks and enabling a compact 3B model to outperform larger open-source variants and proprietary models like GPT-4o. Keywords: Latent Visual Reasoning · Knowledge Distillation · MLLMs 1 Introduction Multimodal Large Language Models (MLLMs) have advanced rapidly in recent years [1, 8, 39], early multimodal reasoning models primarily thinking about images, captioning visual inputs into text and reasoning mainly in the language space via explicit chains of thought [21, 33, 54]. Recent work instead emphasizes thinking with images, more tightly integrating visual evidence into the reasoning process [20,37,59], leading to improved performance on complex visual reasoning tasks [15, 45, 52, 56] Subsequently, latent reasoning [18] has emerged as a complementary direction, compressing intermediate reasoning into continuous hidden states rather than explicit CoT. This paradigm has been extended to MLLMs to model abstract visual thoughts within latent tokens [27,55]. While effective, existing methods largely rely on manually designed visual supervision, such as auxiliary images * Equal contribution. 2 L. Wu, T. Jiang et al. Previous Training Previous Training 1. Identify Visual Contexts （𝑽sem ） 2. Look at A and B （𝑽traj ） 3. Answer：B 1. Answer: B Teacher Model Question: Which point is closer to the camera, Point A