cs.CVNov 18, 2026

The Elephant in the Question: Unified Models Picture the Concepts Behind Their Answers

Authors: Tobias Braun, Marcus Rohrbach, Lorenzo Cavallaro

Organizations: Technische Universität Darmstadt · University College London · Mohamed bin Zayed University of Artificial Intelligence

Abstract

Visualizing intermediate representations could help us inspect the reasoning steps underlying a model output. Unified multimodal models offer this opportunity by combining understanding and generation in one architecture. Their image generators can access representations formed while processing text. We introduce the V-lens, which uses this connection to visualize a chosen token's hidden state without training a decoder. Across four diffusion-based and two purely autoregressive unified models, the resulting images reveal key intermediate concepts that connect a question to its answer even when the concepts are never named explicitly. Controlled tests show how this content changes with context, token position, and depth. We then introduce lens tuning, which supervises selected hidden states with pictures through the V-lens. Across BAGEL, SenseNova-U1, and HunyuanImage-3.0, the rate of correct visualizations of pronoun referents rises from 2-11% to 86-94%. Surprisingly, such visual supervision also improves text answers without supervising the answer tokens directly. Code is available at https://github.com/multimodal-ai-lab/V-lens.

Figures & tables

Appendix figures & tables29 assets

Supplementary material from the paper’s appendix.

Appendix

Explore similar work

Jun 25, 2026

Ask, Solve, Generate: Self-Evolving Unified Multimodal Understanding and Generation via Self-Consistency Rewards

Most unified large multimodal models (LMMs) that support both visual understanding and image generation still rely on curated post-training supervision, such as human annotations, preference labels, or external reward models. We ask whether a unified LMM can improve both abilities autonomously using only unlabeled images. We propose a self-evolving training framework with three internal roles: a Proposer that generates visual questions, a Solver that answers and evaluates them, and a Generator that synthesizes images. Training uses only self-derived consistency signals, without human annotations, preference labels, or task-trained external reward/judge models. To stabilize learning, we introduce Solver Token Entropy (STE), a continuous difficulty signal based on token-level prediction uncertainty that remains useful even when sample-level consistency becomes unreliable. For image generation, we design a multi-scale internal evaluation scheme that combines question-answer fidelity scoring with cycle-consistent captioning. This creates a solver-mediated coupling, where better visual understanding enables more reliable generation assessment and stronger internal training signals. The framework preserves the same role decomposition, reward logic, and training schedule across diffusion-based BLIP3o, rectified-flow BAGEL, and autoregressive VARGPT-v1.1 architectures, requiring only each backbone's native prompting and generation interface. Across eight understanding metrics, our method consistently improves over the corresponding base models. On BAGEL, it achieves a +3.5%+3.5\% absolute gain on MMMU and improves GenEval image generation performance from 82%82\% to 85%85\%. Code and models are publicly released.
May 12, 2026

UniVLR: Unifying Text and Vision in Visual Latent Reasoning for Multimodal LLMs

Multimodal large language models are increasingly expected to perform thinking with images, yet existing visual latent reasoning methods still rely on explicit textual chain-of-thought interleaved with visual latent tokens. This interleaved design limits efficiency and keeps reasoning fragmented across separate text and vision channels. We propose UniVLR, a unified visual latent reasoning framework that treats textual reasoning and auxiliary visual evidence as a shared visual workspace. Instead of preserving text CoT as an independent inference-time path, UniVLR renders reasoning traces together with auxiliary images and learns to compress this unified representation into compact visual latent tokens. At inference time, the model reasons only through visual latents and directly decodes the final answer, avoiding both external tool calls and verbose text reasoning. Experiments on real-world perception and visual reasoning tasks show that UniVLR outperforms prior visual latent reasoning methods while using substantially fewer generated reasoning tokens, suggesting a more unified and efficient paradigm for visual thinking in MLLMs.
Nov 17, 2026

Pixel is All You Need

Text, images, audio, touch, and even robot actions can, in principle, all be rendered onto a single canvas, suggesting an ambitious goal of unifying all modalities within vision. Language provides the natural first test. To subsume it, a visual model must reason directly over text in pixels. Across a wide range of image, video, and unified multimodal generators, however, we find that their visual generation modules cannot perform symbolic reasoning over text on the canvas. They succeed only when an underlying text module or prompt rewriter already supplies the answer, effectively reducing the generator to a mere renderer, a limitation that post-training does little to resolve. To test whether this failure stems from pixels themselves or from current training paradigms, we train a family of purely pixel-space generative models from scratch without any text pathway. With images as their sole input and output, these models read questions, write intermediate steps, and produce answers directly on the canvas. Remarkably, they exhibit hallmark behaviors of large language models. Performance improves predictably with scale following a power law. Writing intermediate steps frame by frame substantially outperforms single-frame answering, echoing chain-of-thought, while reinforcement-learning post-training further strengthens reasoning. Moreover, sharing one canvas allows the same model to reason across modalities, a capability that text-only language models lack. To the best of our knowledge, this is the first systematic study of these phenomena in purely pixel-space generative models free of text tokens.