Vision-Language Models (VLMs) often suffer from visual hallucinations: generating things that are not consistent with visual inputs and language shortcuts, where they skip the visual part and just rely on text priors. These issues arise because most post training methods for VLMs rely on simple verifiable answer matching and supervise only final outputs, leaving intermediate visual reasoning without explicit guidance. As a result, VLMs receive sparse visual signals and often learn to prioritize language based reasoning over visual perception. We introduce Vision SR1, a three stage self rewarding reinforcement learning method that improves visual reasoning without relying on external visual supervision. Vision SR1 decomposes VLM reasoning into two components: visual reasoning and language reasoning, where the model is first prompted to produce self-contained visual descriptions sufficient to answer the question without referring back to the input image, before jointly optimizing both visual and language reasoning through our multi reward loss objective. To validate this self containment, the same VLM model is reprompted to perform language reasoning using only the generated visual reasoning as input to compute visual reward. The final reward is computed through a decoupled reward-advantage framework, where visual reward and language reasoning reward each have their advantages calculated separately. Our experiments show that Vision SR1 improves visual reasoning, mitigates visual hallucinations, and reduces reliance on language shortcuts across diverse vision language tasks, while being more efficient than methods that rely on external visual reward models, which require additional GPUs to host. In contrast, Vision SR1 introduces no extra GPU overhead beyond that of standard training.
Faithful-MR1: Faithful Multimodal Reasoning via Anchoring and Reinforcing Visual Attention
AMAP, Alibaba Group（阿里云实验室，阿里巴巴集团）；University of Chinese Academy of Sciences（中国科学院大学）；Tsinghua University（清华大学）；Nanyang Technological University（南洋理工大学）
VisDoT : Enhancing Visual Reasoning through Human-Like Interpretation Grounding and Decomposition of Thought
Department of Computer Science and Artificial Intelligence, Dongguk University（东国大学计算机科学与人工智能系）；Department of Electronics and Electrical Engineering, Dongguk University（东国大学电子与电气工程系）
GeoAnchor: Collaborative Reasoning via Latent Decomposition for 3D Spatial Understanding
Shanghai Jiao Tong University（上海交通大学）；Xingchen AGI Lab, China Telecom Artificial Intelligence Technology (Beijing) Co., Ltd（星辰通用人工智能实验室，中国电信人工智能技术（北京）有限公司）；University of Science and Technology Beijing（北京科技大学）；The Hong Kong University of Science and Technology (Guangzhou)（香港科技大学（广州））
Can Atomic Step Decomposition Enhance the Self-structured Reasoning of Multimodal Large Models?
Sun Yat-sen University（中山大学）；Hong Kong University of Science and Technology（香港科学与技术大学）；Shanghai Jiaotong University（上海交通大学）；University of Hong Kong（香港大学）；Huawei Noah’s Ark Lab（华为诺亚实验室）
Northwestern Polytechnical University（西北工业大学）；The University of Sydney（悉尼大学）；SenseTime（商汤科技）
Pedestrian counting is a fundamental tool for understanding pedestrian patterns and crowd flow analysis. Existing works (e.g., image-level pedestrian counting, crossline crowd counting et al.) either only focus on the image-level counting or are constrained to the manual annotation of lines. In this work, we propose to conduct the pedestrian counting from a new perspective - Video Individual Counting (VIC), which counts the total number of individual pedestrians in the given video (a person is only counted once). Instead of relying on the Multiple Object Tracking (MOT) techniques, we propose t…
Sun Yat-sen University（中山大学）；Lanzhou University（兰州大学）；University of Hong Kong（香港大学）；Hefei University of Technology（合肥工业大学）；National University of Singapore（新加坡国立大学）
EvoLMM: Self-Evolving Large Multimodal Models with Continuous Rewards
Mohamed bin Zayed University of AI（Mohamed bin Zayed人工智能大学）；Aalto University（阿alto大学）；Australian National University（澳大利亚国立大学）；Linköping University（林肯大学）
Temporal reasoning is an important aspect of video analysis. 3D CNN shows good performance by exploring spatial-temporal features jointly in an unconstrained way, but it also increases the computational cost a lot. Previous works try to reduce the complexity by decoupling the spatial and temporal filters. In this paper, we propose a novel decomposition method that decomposes the feature channels into spatial and temporal groups in parallel. This decomposition can make two groups focus on static and dynamic cues separately. We call this grouped spatial-temporal aggregation (GST). This decomposi…
RADIANT-PET: Reasoning-Augmented PET/CT Lesion Segmentation with Large Language Models and Reinforcement Learning
Division of Hematology, The Ohio State University Comprehensive Cancer Center（血液科，俄亥俄州立大学综合癌症中心）；Department of Radiology, Mahidol University（医学放射科，玛希多大学）；Department of Radiology, Mettapracharak Hospital（医学放射科，Mettapracharak医院）；Department of Radiation Oncology, The Ohio State University Comprehensive Cancer Center（放射肿瘤科，俄亥俄州立大学综合癌症中心）
