The study investigates how text-to-image models interpret number words beyond ten objects. By keeping the object type and initial noise constant, researchers varied the requested count and observed that models reliably follow the number only up to about ten items; beyond that, the count influence weakens and the image composition is largely governed by the coarse noise. Models such as SD3.5‑Medium and FLUX.1‑dev lose adherence after ten, while FLUX.2 and Qwen‑Image loosely track counts up to twenty‑five, yet no clear directional effect is achieved beyond the models’ limits.
By Liang Yue, Eric A. Vance
arXiv:2610.27756v1 Announce Type: new
Abstract: Vision-language models (VLMs) are exact on small sets and drift into estimation on larger ones, a profile reminiscent of children before they master co...
By Yingxin Jia, Hui Geng, Wenxin Yang, Mohan Zhang
arXiv:2604.25213v2 Announce Type: replace
Abstract: OpenAI's GPT-Image-2 has effectively erased the visual boundary between authentic and AI-edited document images: a single number on a receipt can b...
By Jiaqi Wu, Yuchen Zhou, Dennis Tsang Ng, Xingyu Shen, Kidus Zewde, Ankit Raj, Tommy Duong, Simiao Ren
The paper critiques current memorization audits for generative models, arguing that lacking a proper null distribution leads to misleading conclusions. It introduces two exact null tests—one permutation test for whole models and a calibrated test for single images—showing that many previously flagged memorizations disappear under these stricter controls. The authors also propose a scale‑restricted statistic based on the Intersection Euler Characteristic Profile to better detect distinct copied images.
By Sushovan Majhi, Pramita Bagchi
arXiv:2606. 31495v1 Announce Type: new Abstract: We study a single idea across two settings: that a prediction-error signal, computed by a small predictor over the latent space of a frozen encoder, can serve both as a gate on plasticity and as a substrate for metacognition.
By Louis Mouchon
The paper investigates how answer candidates in masked diffusion multimodal large language models (MLLMs) can stabilize before their rationales are fully generated. It distinguishes between retrospective stabilization of the logged candidate and token commitment, and evaluates these two temporal markers across three visual question‑answering benchmarks. The study finds that a large portion of the rationale canvas remains unwritten at stabilization, that reducing block length dramatically lowers this fraction, and that direct prompting instructions can significantly alter accuracy depending on the model and dataset. The authors also use matched‑canvas image ablations to separate visual sensitivity from answer stabilization, concluding that coverage rather than conditional accuracy drives most prompting differences.
By Keuntae Kim, Yong Suk Choi
Scout-Refine-Correct (SRC) is a training recipe that equips flow-matching image generation models with an inference-time refinement capability. During sampling, the model performs a scout rollout to predict a full sample, uses this prediction as a self-conditioning signal, and then takes the actual step; this process can be repeated for multiple refinement rounds. On ImageNet-256, SRC improves LightningDiT-B’s FID from ~14.5 to ~7.26 with 550 normalised flow evaluations, matching a best-of‑N baseline that relies on an oracle verifier, and further reduces FID to 3.81 when combined with best-of‑N.
By Sergey Karpukhin, Aibek Alanov
The study investigates whether a multimodal model like Gemma4-12B-it uses a shared internal concept of numbers across different modalities and operations. By applying causal interventions to the model’s residual stream, the authors identify a small subspace where swapping coordinates changes the model’s arithmetic answer. They find that while the subspaces for text and image are not identical, a single rotation and rescaling aligns them, revealing a shared geometric structure that encodes numbers in hidden directions used to generate multi‑digit outputs over time.
By Oriol Juan Sabater, Alex Ferrando de las Morenas, Pau Rodr\'iguez, Xavier Suau
The paper introduces Shift‑Reliance Alignment (SRA), a method for ranking vision‑language models (VLMs) based on how much they rely on specific cues such as scene context or text. By editing source images to mislead these cues and measuring model responses, SRA predicts which VLMs will be misled by physical sticker attacks and which will hallucinate on unedited photographs. In pre‑registered tests, SRA selects models that reduce error on challenging datasets like Objects365 and OpenImages, outperforming selection by source error and model confidence on the CueDial benchmark.
By Faisal Mohammad, Mohamed Younis
arXiv:2609.15635v1 Announce Type: cross
Abstract: A radiology report can already answer a clinical question, so it is hard to tell whether a vision-language model also uses the image. ModaLens, a pai...
By Sebasti\'an Andr\'es Cajas Ord\'o\~nez, Maximin Lange, Quang Bui, Anqi Peter Li, Felipe Ocampo Osorio, Rafi Al Attrach, Kushul Reddy Palakala, Sahil Kapadia, Zakaria Laouabdia Sellami, Xinyue Zhang, Ashley Zhang, Leo Anthony Celi
arXiv:2609.09528v1 Announce Type: new
Abstract: Video large language models (Video-LLMs) are increasingly used as the perceptual front end of world models, a role that assumes they can read motion: h...
By Dhairya Bhatia, Bishoy Galoaa, Oliver Fritsche, Shahid Kamal, Muhammad Obaidullah Abdul Salam, Umer Saleem, Om Rastogi, Frania Felix Chettiar, Nesli Erdogmus, Sarah Ostadabbas
arXiv:2609.13308v1 Announce Type: cross
Abstract: A companion evaluation found that naming the target part in a manipulation prompt increased action accuracy by 0.32-0.63 across eight vision-language...
By Sarthak Sattigeri