MagiC is a benchmark for grounded multimodal cognition that evaluates vision‑language models on answer accuracy, step‑by‑step reasoning quality, grounding fidelity, and self‑correction. It contains about 5,500 weakly supervised QA examples generated from strong model outputs and 900 human‑curated examples with fine‑grained annotations, including rationales and bounding box groundings. The benchmark tests 15 models ranging from 7B to 70B parameters, introduces new metrics such as MagiScore and StepSense, and includes diagnostic settings to probe robustness against adversarial visual cues and introspective error correction.
By Chengfei Wu, Ronald Seoh, Bingxuan Li, Liqiang Zhang, Fengrong Han, Dan Goldwasser
The paper introduces the Very Big Video Reasoning (VBVR) Dataset, a large-scale collection of over one million video clips organized into 200 curated reasoning tasks. It also presents VBVR-Bench, a benchmark framework that uses rule-based, human-aligned scorers for reproducible evaluation of video reasoning models. The authors conduct a large-scale scaling study, noting early signs of emergent generalization to unseen reasoning tasks, and make all resources publicly available.
By Maijunxian Wang, Ruisi Wang, Juyi Lin, Ran Ji, Thadd\"aus Wiedemer, Qingying Gao, Dezhi Luo, Yaoyao Qian, Lianyu Huang, Zelong Hong, Jiahui Ge, Qianli Ma, Hang He, Yifan Zhou, Lingzi Guo, Lantao Mei, Jiachen Li, Hanwen Xing, Tianqi Zhao, Fengyuan Yu, Weihang Xiao, Yizheng Jiao, Jianheng Hou, Danyang Zhang, Pengcheng Xu, Boyang Zhong, Zehong Zhao, Gaoyun Fang, John Kitaoka, Yile Xu, Hua Xu, Kenton Blacutt, Tin Nguyen, Siyuan Song, Haoran Sun, Shaoyue Wen, Linyang He, Runming Wang, Yanzhi Wang, Mengyue Yang, Ziqiao Ma, Rapha\"el Milli\`ere, Freda Shi, Nuno Vasconcelos, Daniel Khashabi, Alan Yuille, Yilun Du, Ziming Liu, Bo Li, Dahua Lin, Ziwei Liu, Vikash Kumar, Yijiang Li, Lei Yang, Zhongang Cai, Hokin Deng
The paper introduces VWG-Bench, a benchmark covering nine reasoning dimensions and 38 tasks to evaluate video generative models on symbolic reasoning, physical laws, and goal pursuit. It also presents Vid-PRE, a prompt-rewriting framework that offloads reasoning to a VLM, improving logical performance without changing the generator architecture. Experiments show that current models excel at visual quality but struggle with logic-heavy tasks, while Vid-PRE significantly boosts reasoning across multiple generators.
By Meng Luo, Yicheng Liu, Jiahao Wang, Yuanxing Zhang, Xin Tao, Pengfei Wan, Kun Gai, Hao Fei
The paper introduces three new vision‑centric evaluation benchmarks—temporal frame retrieval, video future prediction, and causal memory distortion—to assess visual question answering in large video models. Unlike traditional benchmarks that rely on text-based multiple choice questions, these tasks require models to reason directly from visual inputs. The authors find that current state‑of‑the‑art models struggle with visual queries, highlighting a gap in visual understanding that future research should address.
By Rwiddhi Chakraborty (Oliver), Yinong (Oliver), Wang, Cheng Zhang, Fan Bai, Zhuoran You, Michael Kampffmeyer, Yong Jae Lee, Fernando De la Torre, Robert Jenssen
arXiv:2601. 07761v2 Announce Type: replace Abstract: Large Vision-Language Models (LVLMs) face a fundamental dilemma in video reasoning: they are caught between the prohibitive computational costs of verbose reasoning and the hallucination risks of efficient, ungrounded approaches.
By Yanxiang Huang, Guohua Gao, Zhaoyang Wei
The paper introduces CAIT, a benchmark of 400 synthetic scenes featuring counter‑intuitive actions that challenge multimodal large language models (MLLMs). Human participants and proprietary models like Claude and Gemini perform well, but standard open‑source instruction‑tuned MLLMs fail, largely due to a strong language prior that overrides contradictory visual evidence. The study shows that Chain‑of‑Thought reasoning can help but introduces new issues, while targeted fine‑tuning and structured prompting can reduce reliance on language priors and improve visual grounding.
By Chen Ling, Tongwei Zhang, Hanqian Li, Nai Ding
Hob‑VL is a benchmark for visually grounded Boolean reasoning that tests models on two tasks: determining whether a Boolean rule holds in an image and identifying the unique object that satisfies a Boolean description. It contains 6,000 balanced Yes/No questions built from ten visual statements across 1,000 generated scenes and 46 labeled photographs, plus 1,000 object‑identification questions on the same photographs. The questions are designed to challenge reasoning with misleading local cues, nested logical operations, and both symbolic and natural‑language presentations, revealing significant gaps in current model performance.
By Yuzhou Wang, Emile Anand, Ijay Narang
CoWorld Reasoner is an agentic framework that leverages a Video World Model (VWM) as a unified visual reasoning engine for Vision‑Language Models (VLMs). It exposes multiple visual operations through tools and skills, allowing the VLM to decide when to imagine, select an operation, generate visual content, and interpret the resulting evidence. The framework includes a VLM‑VWM Harness Co‑Evolution mechanism that jointly refines tools, skills, and execution code based on feedback, while keeping model weights fixed. Across seven benchmarks covering spatial, temporal, causal, and perceptual reasoning, CoWorld Reasoner outperforms direct VLM reasoning and improves accuracy on all benchmarks, achieving notable gains on MMSI and ERQA.
By Dongjie Cheng, Ruifeng Yuan, Ran Wei, Ning Song, Yongqi Li, Wenjie Li
arXiv:2603.14733v2 Announce Type: replace
Abstract: Multimodal Large Language Models have achieved strong performance in single-video understanding, yet their ability to reason across multiple videos...
By Yue Zhang, Liqiang Jing, Jia Li, Yapeng Tian, Xinya Du, Yunhui Guo, Vibhav Gogate
SUREON is a new video question‑answering dataset and benchmark that tests visually grounded surgical reasoning. It contains 76K training pairs from narrated surgical videos and a 340‑question test set verified by expert surgeons, designed to prevent models from relying on language shortcuts. Baseline models trained on SUREON outperform existing open‑source models and show that surgical visual reasoning remains largely unsolved, with no model exceeding 53% accuracy.
By Alejandra Perez, Anita Rau, Busisiwe Mlambo, Lee White, Aina Brunet, Juan Verde, Juan Caicedo, Jo\"el L. Lavanchy, Adam Schmidt, Chinedu Nwoye, Omid Mohareri, Muhammad Abdullah Jamal
EDCT-Bench is a benchmark that evaluates the faithfulness of Vision‑Language Models (VLMs) by using Explanation‑Driven Counterfactual Testing (EDCT). EDCT extracts visual concepts from a model’s natural language explanation, applies minimal verified edits to those concepts, and checks whether the model’s answer and explanation remain consistent with the edited image. The benchmark covers three domains—knowledge‑intensive VQA, safety‑critical driving, and 3D spatial reasoning—and reveals significant faithfulness gaps in current VLMs, while also showing that EDCT‑generated counterfactuals can improve training.
By Sihao Ding, Santosh Vasa, Aditi Ramadwar, Thomas Monninger
arXiv:2506.09557v2 Announce Type: replace-cross
Abstract: While Multimodal Large Language Models (MLLMs) demonstrate impressive performance in benign scenarios, their cognitive reliability deteriorat...
By Zhaoyang Wei, Bowen Jiang, Xumeng Han, Jiashu Li, Xuehui Yu, Yuling Liu, Guorong Li, Zhenjun Han, Jianbin Jiao