arXiv Machine Learning
Aug 27

VBVR-Pro: A Scalable and Verifiable Suite for Native Visual Reasoning

VBVR-Pro is a closed‑loop testbed that enables native visual reasoning through generation, offering 300 procedurally generated tasks that scale training and allow strong transfer to external benchmarks. It supplies verifiable reward scorers based on deterministic, task‑specific rules, outperforming VLM‑as‑a‑judge approaches and providing reliable signals for reinforcement learning. The suite also facilitates controlled modality studies, revealing that video generation excels at persistent spatiotemporal tracking while interleaved generation offers a compute‑efficient alternative, and highlights the importance of vision‑native trajectories for reasoning.

By Junxiang Xu, Ruisi Wang, Fanyi Pu, Maijunxian Wang, Ran Ji, Tongxi Zhou, Chenyang Gu, Jing Zuo, Hongcan Xiao, Yimeng Geng, Wanqi Yin, Wei Chen, Oscar Qian, Zhengan Yan, Ziqi Huang, Haiwen Diao, Liang Pan, Bo Li, Xiangyu Fan, Dezhi Luo, Fengyuan Yu, Zehong Zhao, Qingying Gao, Tinghui Zhu, Yilan Zhang, Jingqi Tong, Pinyuan Feng, Zhengze Jiang, Letian Wang, Ziyu Guo, Renrui Zhang, Jieneng Chen, Sonia Joseph, Constantin Venhoff, Saman Motamed, Mengyue Yang, Chandra Sripada, Alan Yuille, Philip Torr, Lvmin Zhang, Vikash Kumar, Daniel Khashabi, Nikolaus Kriegeskorte, Rapha\"el Milli\`ere, Vincent C. M\"uller, Anyi Rao, Quan Wang, Ziwei Liu, Dahua Lin, Lei Yang, Hokin Deng, Zhongang Cai
arXiv Computation and Language
Sep 11

OmniHallu: Unified Hallucination Detection for Cross-Modal Comprehension and Generation in Multimodal Large Language Models

OmniHallu is a unified framework for detecting hallucinations in multimodal large language models across both comprehension and generation tasks involving image, video, and audio modalities. It introduces OmniHallu-Bench, a 10,000-sample benchmark with claim-level human annotations for six cross-modal tasks (I2T, V2T, A2T, T2I, T2V, T2A). The system uses a multi‑agent architecture that decomposes outputs into atomic claims, verifies them with modality‑specific experts, and aggregates evidence through structured reasoning, while a preference‑optimized verifier reduces expert calls by 66% with minimal performance loss.

By Jianjiang Yang, Peihang Li, Shanqing Xu, Mengchen Qian, Lu Zhang, Meng Luo
arXiv Computer Vision
Sep 14

Unified Text-Image Generation with Weakness-Targeted Post-Training

The paper introduces a post‑training approach that enables a single inference process to transition from text reasoning to image synthesis, eliminating the need for explicit modality switching. Using the 14B BAGEL model, the authors demonstrate that targeted post‑training data and reward‑weighted training improve multimodal image generation across four independent T2I benchmarks. The study highlights the benefits of joint text‑image generation and strategic data selection for enhancing T2I performance.

By Jiahui Chen, Philippe Hansen-Estruch, Xiaochuang Han, Yushi Hu, Emily Dinan, Amita Kamath, Michal Drozdzal, Reyhane Askari-Hemmat, Luke Zettlemoyer, Marjan Ghazvininejad
arXiv AI
Oct 27

Can Multi-modal (reasoning) LLMs work as deepfake detectors?

The paper investigates whether state‑of‑the‑art multi‑modal large language models (LLMs) can detect deepfake images. Twelve recent LLMs were benchmarked against traditional detectors on multiple datasets, including real‑world deepfakes, and prompt tuning was used to improve performance. Results show that the best LLMs can match or exceed conventional methods on out‑of‑distribution data, though many LLMs perform poorly; larger model size helps, but newer versions and reasoning abilities do not necessarily improve detection.

By Simiao Ren (Dennis), Yao Yao (Dennis), Kidus Zewde (Dennis), Zisheng Liang (Dennis), Tsang (Dennis), Ng (Dennis), Ning-Yau Cheng, Xiaoou Zhan, Qinzhe Liu, Yifei Chen, Hengwei Xu