# The Expense of Seeing: Attaining Trustworthy Multimodal Reasoning Within the Monolithic Paradigm

arXiv: https://arxiv.org/abs/2604.20665 · HTML: https://mlsanity.com/paper/2604.20665/

## Key facts

- Cited by 2 papers in the map (data through 14 Oct 2026).
- Placed in Multimodal reasoning, one of 17 directions in LLM reasoning.
- Builds on 4 of its 16 references that are in the map.
- First posted on arXiv 22 Apr 2026.

## Placed in

- LLM reasoning / [Multimodal reasoning](https://mlsanity.com/lr/multimodal-reasoning/)
- Multimodal LLMs / vision-language models / [VLM benchmarks and visual reasoning](https://mlsanity.com/mllm/vlm-benchmarks-and-visual-reasoning/)

## Builds on (most cited first, 4 of 4)

- [An Introduction to Vision-Language Modeling](https://mlsanity.com/paper/2405.17247/) (2024)
- [BabyVision: Visual Reasoning Beyond Language](https://mlsanity.com/paper/2601.06521/) (2026)
- [Explain Before You Answer: A Survey on Compositional Visual Reasoning](https://mlsanity.com/paper/2508.17298/) (2025)
- [Improving Fine-grained Visual Understanding in VLMs through Text-Only Training](https://mlsanity.com/paper/2412.12940/) (2024)

## Built on by (most cited first, 2 of 2)

- [ENTRAP-VL: A Taxonomic Probe for Dual Contextual Entrainment in Vision-Language Models](https://mlsanity.com/paper/2607.20092/) (2026)
- [When Can Text Replace Vision? Structural Bottlenecks in Diagram Reasoning](https://mlsanity.com/paper/2609.39142/) (2026)

## Abstract (arXiv)

The rapid proliferation of Vision-Language Models (VLMs) is often framed as enabling unified multimodal knowledge discovery but rests on an under-examined assumption: that current VLMs faithfully synthesise multimodal data. We argue they often do not, and this gap reflects a trustworthiness problem in the dominant Vision Encoder-Projector-LLM paradigm. Rather than extracting grounded knowledge from visual inputs, state-of-the-art models frequently exhibit functional blindness, i.e., exploiting strong language priors to bypass severe visual representation bottlenecks. In this work, we challenge the conventional methodology of multimodal evaluation, which relies on data ablation or new dataset creation and therefore conflates dataset biases with architectural incapacity. We propose an information-theoretic departure: the Modality Translation Protocol, designed to quantify what we call the Expense of Seeing. By translating semantic payloads rather than ablating them, we formulate three novel metrics -- the Toll (ToS), Curse (CoS), and Fallacy (FoS) of Seeing -- culminating in the Semantic Sufficiency Criterion (SSC). Furthermore, we hypothesise a Divergence Law of Multimodal Scaling: as the underlying language engines scale to unprecedented reasoning capabilities, the penalty of the visual knowledge bottleneck may increase rather than diminish. We argue the community should move beyond "multimodal gain" as a primary evaluation target. By elevating the SSC from a passive diagnostic constraint to an active architectural blueprint, we provide a foundation for guiding the next generation of AI systems toward genuine multimodal reasoning.