Kangrui Ruan

Xupeng Chen, Zhixin Lai, Kangrui Ruan et al.

Artificial intelligence has made significant strides in medical visual question answering (Med-VQA), yet prevalent studies often interpret images holistically, overlooking the visual regions of interest that may contain crucial information, potentially aligning with a doctor's prior knowledge that can be incorporated with minimal annotations (e.g., bounding boxes). To address this gap, this paper introduces R-LLaVA, designed to enhance biomedical VQA understanding by integrating simple medical annotations as prior knowledge directly into the image space through CLIP. These annotated visual regions of interest are then fed into the LLaVA model during training, aiming to enrich the model's understanding of biomedical queries. Experimental evaluation on four standard Med-VQA datasets demonstrates R-LLaVA's superiority over existing state-of-the-art (SoTA) methods. Additionally, to verify the model's capability in visual comprehension, a novel multiple-choice medical visual understanding dataset is introduced, confirming the positive impact of focusing on visual regions of interest in advancing biomedical VQA understanding.

Kangrui Ruan, Xinyang Wang, Xuan Di

Social media has become an important platform for people to express their opinions towards transportation services and infrastructure, which holds the potential for researchers to gain a deeper understanding of individuals' travel choices, for transportation operators to improve service quality, and for policymakers to regulate mobility services. A significant challenge, however, lies in the unstructured nature of social media data. In other words, textual data like social media is not labeled, and large-scale manual annotations are cost-prohibitive. In this study, we introduce a novel methodological framework utilizing Large Language Models (LLMs) to infer the mentioned travel modes from social media posts, and reason people's attitudes toward the associated travel mode, without the need for manual annotation. We compare different LLMs along with various prompting engineering methods in light of human assessment and LLM verification. We find that most social media posts manifest negative rather than positive sentiments. We thus identify the contributing factors to these negative posts and, accordingly, propose recommendations to traffic operators and policymakers.

Yifeng Ding, Hung Le, Songyang Han et al.

Training Large Language Models (LLMs) for multi-turn Tool-Integrated Reasoning (TIR) - where models iteratively reason, generate code, and verify through execution - remains challenging for existing reinforcement learning (RL) approaches. Current RL methods, exemplified by Group Relative Policy Optimization (GRPO), suffer from coarse-grained, trajectory-level rewards that provide insufficient learning signals for complex multi-turn interactions, leading to training stagnation. To address this issue, we propose Group Turn Policy Optimization (GTPO), a novel RL algorithm specifically designed for training LLMs on multi-turn TIR tasks. GTPO introduces three key innovations: (1) turn-level reward assignment that provides fine-grained feedback for individual turns, (2) return-based advantage estimation where normalized discounted returns are calculated as advantages, and (3) self-supervised reward shaping that exploits self-supervision signals from generated code to densify sparse binary outcome-based rewards. Our comprehensive evaluation demonstrates that GTPO outperforms GRPO by 3.0% on average across diverse reasoning benchmarks, establishing its effectiveness for advancing complex mathematical reasoning in the real world.

Mobin Zhao, Wangzhi Li, Yongjie Fu et al.

This paper develops a decentralized reinforcement learning (RL) scheme for multi-intersection adaptive traffic signal control (TSC), called "CVLight", that leverages data collected from connected vehicles (CVs). The state and reward design facilitates coordination among agents and considers travel delays collected by CVs. A novel algorithm, Asymmetric Advantage Actor-critic (Asym-A2C), is proposed where both CV and non-CV information is used to train the critic network, while only CV information is used to execute optimal signal timing. Comprehensive experiments show the superiority of CVLight over state-of-the-art algorithms under a 2-by-2 synthetic road network with various traffic demand patterns and penetration rates. The learned policy is then visualized to further demonstrate the advantage of Asym-A2C. A pre-train technique is applied to improve the scalability of CVLight, which significantly shortens the training time and shows the advantage in performance under a 5-by-5 road network. A case study is performed on a 2-by-2 road network located in State College, Pennsylvania, USA, to further demonstrate the effectiveness of the proposed algorithm under real-world scenarios. Compared to other baseline models, the trained CVLight agent can efficiently control multiple intersections solely based on CV data and achieve the best performance, especially under low CV penetration rates.