Rui Hao

Rui Hao, Linmei Hu, Weijian Qi et al.

Dialogue-based language models mark a huge milestone in the field of artificial intelligence, by their impressive ability to interact with users, as well as a series of challenging tasks prompted by customized instructions. However, the prevalent large-scale dialogue-based language models like ChatGPT still have room for improvement, such as unstable responses to questions and the inability to think cooperatively like humans. Considering the ability of dialogue-based language models in conversation and their inherent randomness in thinking, we propose ChatLLM network that allows multiple dialogue-based language models to interact, provide feedback, and think together. We design the network of ChatLLMs based on ChatGPT. Specifically, individual instances of ChatGPT may possess distinct perspectives towards the same problem, and by consolidating these diverse viewpoints via a separate ChatGPT, the ChatLLM network system can conduct decision-making more objectively and comprehensively. In addition, a language-based feedback mechanism comparable to backpropagation is devised to update the ChatGPTs within the network. Experiments on two datasets demonstrate that our network attains significant improvements in problem-solving, leading to observable progress amongst each member.

Jiping Luo, Tingting Zhang, Rui Hao et al.

Cooperative coordination at unsignalized road intersections, which aims to improve the driving safety and traffic throughput for connected and automated vehicles, has attracted increasing interests in recent years. However, most existing investigations either suffer from computational complexity or cannot harness the full potential of the road infrastructure. To this end, we first present a dedicated intersection coordination framework, where the involved vehicles hand over their control authorities and follow instructions from a centralized coordinator. Then a unified cooperative trajectory optimization problem will be formulated to maximize the traffic throughput while ensuring the driving safety and long-term stability of the coordination system. To address the key computational challenges in the real-world deployment, we reformulate this non-convex sequential decision problem into a model-free Markov Decision Process (MDP) and tackle it by devising a Twin Delayed Deep Deterministic Policy Gradient (TD3)-based strategy in the deep reinforcement learning (DRL) framework. Simulation and practical experiments show that the proposed strategy could achieve near-optimal performance in sub-static coordination scenarios and significantly improve the traffic throughput in the realistic continuous traffic flow. The most remarkable advantage is that our strategy could reduce the time complexity of computation to milliseconds, and is shown scalable when the road lanes increase.

Qixing Zhou, Jiacheng Zhang, Haiyang Wang et al.

Agents powered by large language models (LLMs) are increasingly adopted in the software industry, contributing code as collaborators or even autonomous developers. As their presence grows, it becomes important to assess the current boundaries of their coding abilities. Existing agentic coding benchmarks, however, cover a limited task scope, e.g., bug fixing within a single pull request (PR), and often rely on non-executable evaluations or lack an automated approach for continually updating the evaluation coverage. To address such issues, we propose FeatureBench, a benchmark designed to evaluate agentic coding performance in end-to-end, feature-oriented software development. FeatureBench incorporates an execution-based evaluation protocol and a scalable test-driven method that automatically derives tasks from code repositories with minimal human effort. By tracing from unit tests along a dependency graph, our approach can identify feature-level coding tasks spanning multiple commits and PRs scattered across the development timeline, while ensuring the proper functioning of other features after the separation. Using this framework, we curated 200 challenging evaluation tasks and 3825 executable environments from 24 open-source repositories in the first version of our benchmark. Empirical evaluation reveals that the state-of-the-art agentic model, such as Claude 4.5 Opus, which achieves a 74.4% resolved rate on SWE-bench, succeeds on only 11.0% of tasks, opening new opportunities for advancing agentic coding. Moreover, benefiting from our automated task collection toolkit, FeatureBench can be easily scaled and updated over time to mitigate data leakage. The inherent verifiability of constructed environments also makes our method potentially valuable for agent training.

Dingbang Wu, Rui Hao, Haiyang Wang et al.

We present MobileGym, a browser-hosted, lightweight, fully controllable environment for everyday mobile use, targeting interaction fidelity without replicating proprietary backends. It enables two capabilities previously out of reach for everyday apps: verifiable outcome signals through deterministic state-based judging over structured JSON state, and scalable online RL through low-cost parallel rollouts. The full environment state is captured, configured, forked, and compared as structured JSON, and a single server can host hundreds of parallel instances, with about 400 MB memory per instance and about 3 s cold start. A layered state model and a declarative task-definition framework keep state programmability and task creation practical at scale, and a single programmatic judging mechanism delivers both deterministic evaluation verdicts and dense RL rewards. The accompanying MobileGym-Bench provides 416 parameterized task templates, including 256 test and 160 train templates, over 28 apps, with deterministic judges and a structured AnswerSheet protocol that avoids free-text matching failures. In a Sim-to-Real case study, GRPO on Qwen3-VL-4B-Instruct gains +12.8 percentage points on the 256-task test set, and on a 59-task real-device signal subset, real-device execution retains 95.1% of the simulation-side training gain. Project page: https://mobilegym.github.io.

Kaiyang Wan, Honglin Mu, Rui Hao et al.

Like humans, Large Language Models (LLMs) struggle to generate high-quality long-form text that adheres to strict requirements in a single pass. This challenge is unsurprising, as successful human writing, according to the Cognitive Writing Theory, is a complex cognitive process involving iterative planning, translating, reviewing, and monitoring. Motivated by these cognitive principles, we aim to equip LLMs with human-like cognitive writing capabilities through CogWriter, a novel training-free framework that transforms LLM constrained long-form text generation into a systematic cognitive writing paradigm. Our framework consists of two key modules: (1) a Planning Agent that performs hierarchical planning to decompose the task, and (2) multiple Generation Agents that execute these plans in parallel. The system maintains quality via continuous monitoring and reviewing mechanisms, which evaluate outputs against specified requirements and trigger necessary revisions. CogWriter demonstrates exceptional performance on LongGenBench, a benchmark for complex constrained long-form text generation. Even when using Qwen-2.5-14B as its backbone, CogWriter surpasses GPT-4o by 22% in complex instruction completion accuracy while reliably generating texts exceeding 10,000 words. We hope this cognitive science-inspired approach provides a paradigm for LLM writing advancements: \href{https://github.com/KaiyangWan/CogWriter}{CogWriter}.

Rui Hao, Dayu Tan, Qiankun Li et al.

Accurate segmentation of organs-at-risk (OARs) is vital for safe and precise radiotherapy and surgery. Most existing studies segment only a limited set of organs or regions, lacking a systematic treatment of OARs segmentation. We present a Multi-scale Cascaded Fusion Network (MCFNet) that aggregates features across multiple scales and resolutions. MCFNet consists of a Sharp Extraction Backbone for the downsampling path and a Flexible Connection Backbone for skip-connection fusion, strengthening representation learning in both stages. This design improves boundary localization and preserves fine structures while maintaining computational efficiency, enabling reliable performance even on low-resolution inputs. Experiments on an NVIDIA A6000 GPU using 36,131 image-mask pairs from 671 patients across 10 datasets show consistent robustness and strong cross-dataset generalization. An adaptive loss-aggregation strategy further stabilizes optimization and yields additional gains in accuracy and training efficiency. Through extensive validation, MCFNet outperforms existing methods, excelling in organ segmentation and providing reliable image-guided support for computer-aided diagnosis. Our solution aims to improve the precision and safety of radiotherapy and surgery while supporting personalized treatment, advancing modern medical technology. The code has been made available on GitHub: https://github.com/Henry991115/MCFNet.

Xinyao Zhang, Rui Wang, Jinhao Cui et al.

Multi-window mobile scenarios, such as split-screen and foldable modes, make GUI display defects more likely by forcing applications to adapt to changing window sizes and dynamic layout reflow. Existing detection techniques are limited in two ways: they are largely passive, analyzing screenshots only after problematic states have been reached, and they are mainly designed for conventional full-screen interfaces, making them less effective in multi-window settings.We propose an end-to-end framework for GUI display defect detection in multi-window mobile scenarios. The framework proactively triggers split-screen, foldable, and window-transition states during app exploration, uses Set-of-Mark (SoM) to align screenshots with widget-level interface elements, and leverages multimodal large language models with chain-of-thought prompting to detect, localize, and explain display defects. We also construct a benchmark of GUI display defects using 50 real-world Android applications.Experimental results show that multi-window settings substantially increase the exposure of layout-related defects, with text truncation increasing by 184% compared with conventional full-screen settings. At the application level, our method detects 40 defect-prone apps with a false positive rate of 10.00% and a false negative rate of 11.11%, outperforming OwlEye and YOLO-based baselines. At the fine-grained level, it achieves the best F1 score of 87.2% for widget occlusion detection.

Jiexin Ding, Bowen Zhao, Yuntao Wang et al.

English as a Second Language (ESL) learners often encounter unknown words that hinder their text comprehension. Automatically detecting these words as users read can enable computing systems to provide just-in-time definitions, synonyms, or contextual explanations, thereby helping users learn vocabulary in a natural and seamless manner. This paper presents EyeLingo, a transformer-based machine learning method that predicts the probability of unknown words based on text content and eye gaze trajectory in real time with high accuracy. A 20-participant user study revealed that our method can achieve an accuracy of 97.6%, and an F1-score of 71.1%. We implemented a real-time reading assistance prototype to show the effectiveness of EyeLingo. The user study shows improvement in willingness to use and usefulness compared to baseline methods.

Rui Hao, Dianbo Liu, Linmei Hu

The merging of human intelligence and artificial intelligence has long been a subject of interest in both science fiction and academia. In this paper, we introduce a novel concept in Human-AI interaction called Symbiotic Artificial Intelligence with Shared Sensory Experiences (SAISSE), which aims to establish a mutually beneficial relationship between AI systems and human users through shared sensory experiences. By integrating multiple sensory input channels and processing human experiences, SAISSE fosters a strong human-AI bond, enabling AI systems to learn from and adapt to individual users, providing personalized support, assistance, and enhancement. Furthermore, we discuss the incorporation of memory storage units for long-term growth and development of both the AI system and its human user. As we address user privacy and ethical guidelines for responsible AI-human symbiosis, we also explore potential biases and inequalities in AI-human symbiosis and propose strategies to mitigate these challenges. Our research aims to provide a comprehensive understanding of the SAISSE concept and its potential to effectively support and enhance individual human users through symbiotic AI systems. This position article aims at discussing poteintial AI-human interaction related topics within the scientific community, rather than providing experimental or theoretical results.

Xincheng He, Lei Xu, Xiangyu Zhang et al.

API recommendation in real-time is challenging for dynamic languages like Python. Many existing API recommendation techniques are highly effective, but they mainly support static languages. A few Python IDEs provide API recommendation functionalities based on type inference and training on a large corpus of Python libraries and third-party libraries. As such, they may fail to recommend or make poor recommendations when type information is missing or target APIs are project-specific. In this paper, we propose a novel approach, PyART, to recommend APIs for Python programs in real-time. It features a light-weight analysis to derives so-called optimistic data-flow, which is neither sound nor complete, but simulates the local data-flow information humans can derive. It extracts three kinds of features: data-flow, token similarity, and token co-occurrence, in the context of the program point where a recommendation is solicited. A predictive model is trained on these features using the Random Forest algorithm. Evaluation on 8 popular Python projects demonstrates that PyART can provide effective API recommendations. When historic commits can be leveraged, which is the target scenario of a state-of-the-art tool ARIREC, our average top-1 accuracy is over 50% and average top-10 accuracy over 70%, outperforming APIREC and Intellicode (i.e., the recommendation component in Visual Studio) by 28.48%-39.05% for top-1 accuracy and 24.41%-30.49% for top-10 accuracy. In other applications such as when historic comments are not available and cross-project recommendation, PyART also shows better overall performance. The time to make a recommendation is less than a second on average, satisfying the real-time requirement.