Aug 2026· arXiv.org· Vol abs/2608.11292· 2 citations· 43 references
Computer Science
TL;DR
The Code-with-Image Bench (CwI-Bench), thirty task families induced by hidden visual computations with disjoint learning and evaluation splits, even GPT-5.6-luna stays below 30% with tool-free chain of thought; given a bare interpreter it reaches 43%, and with skills evolved through its own executable reflection, 67%.
Abstract
Multimodal models increasingly reach for tools when solving visual tasks (crop, zoom, rotate, brighten), a paradigm known as thinking-with-images. The central challenge is one of perception: tools mostly serve to expose visual evidence, reasoning over that evidence stays in language, and most targets are ones a human could in principle determine by inspection. Some visual questions, however, are not bottlenecked by perception: recovering their answers requires executing a multi-step visual algorithm over the pixels. On such questions a model often names the correct algorithm at once yet still answers wrong, because language can describe an algorithm without being able to run one. Code-with-Image crosses that line: given nothing but a Python interpreter, the model must implement a genuine visual algorithm in code to solve the task; the program itself becomes the reasoning. The bottleneck then shifts from executing code to deciding which algorithm to implement. So we let the model teach itself: a training-free reflection loop studies its own failed programs, tests repairs against constructive ground truth, and keeps what survives as portable skills. On our Code-with-Image Bench (CwI-Bench), thirty task families induced by hidden visual computations with disjoint learning and evaluation splits, even GPT-5.6-luna stays below 30% with tool-free chain of thought; given a bare interpreter it reaches 43%, and with skills evolved through its own executable reflection, 67%. The open 27B model climbs the same ladder (9% $\rightarrow$ 33% $\rightarrow$ 56%), and the skills are plain text, transferable across scales and families. When code carries the reasoning, debugging code becomes debugging reasoning.
Across six diverse visual reasoning tasks including multi-view spatial reasoning and collision prediction, ReImaGin consistently outperforms both text-only reasoning and specialist vision-tool baselines, with gains of up to 25\%, demonstrating the advantage of flexible, generative visual reasoning.
Nishad Singhi, Hector P. Rodriguez, Aditya Arora et al.· arXiv.org· 0 citations
Vision-language models (VLMs) have advanced rapidly in multimodal reasoning, yet they remain limited on tasks requiring fine-grained understanding of small or easily overlooked image regions. Recent thinking with images approaches mitigate this limitation by allowing models to inspect images through operations such as...
Ze-Rui Chen, Chang-Rui Chen, Fei Ni et al.· 0 citations
Visual counting benchmarks report whether vision-language models produce the correct number, but not whether errors arise from perception or reasoning. We introduce DecoyCount, a benchmark built from 1,586 real-world images with designed decoys across seven mechanisms, along with paired prompts: a neutral question, a m...
Qiu-Yi Zhang, Zhen Xu, Jing Gu et al.· 0 citations
We present See, Think, Use, a protocol that traces each answer of a vision-language model to perception, reasoning or the use of perception. The diagnosis shows a developer which stage to improve and directs test-time compute to the stage that fails. Benchmarks score the final answer, and an answer that rests on the im...
Jia-Cheng Ding, Yan-Liang Qi, Cong-Yu Guo et al.· 0 citations
MM-OPD is introduced, a multimodal on-policy self-distillation framework for symbolic-to-visual correction that transfers guidance from symbolic-conditioned behavior to the image-conditioned policy through residual token-level targets, steering the model toward correct visual evidence.
Vision-language models can get individual judgments right while struggling to recover and consistently use the exact properties of a 3D object. To characterize this gap, we introduce BedBench, a controlled benchmark of thirteen procedurally generated tasks spanning individual judgments and composite answers. Across twe...
Khai Le-Duc, Nguyen Kim Hai Bui, An Ngo et al.· 1 citation
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.