Author name cluster

Xiaorui Wang

Possible papers associated with this exact author name in Arrow. This page groups case-insensitive exact name matches and is not a full identity disambiguation profile.

8 papers

2 author rows

AAAI Conference 2026 Conference Paper

MCP-AgentBench: Evaluating Real-World Language Agent Performance with MCP-Mediated Tools

Zikang Guo
Benfeng Xu
Chiwei Zhu
Wentao Hong
Xiaorui Wang
Zhendong Mao

The Model Context Protocol (MCP) is rapidly emerging as a pivotal open standard, designed to enhance agent-tool integration and interoperability, and is positioned to unlock a new era of powerful, interconnected, and genuinely utilitarian agentic AI. However, despite MCP's growing adoption, existing benchmarks often fail to capture real-world agent performance within this new paradigm, leading to a distorted perception of their true operational value and an inability to reliably differentiate proficiencies. To bridge this critical evaluation gap, we introduce MCP-AgentBench—a comprehensive benchmark specifically engineered to rigorously assess language agent capabilities in MCP-mediated tool interactions. Core contributions of MCP-AgentBench include: the establishment of a robust MCP testbed comprising 33 operational servers with 188 distinct tools; the development of a benchmark featuring 600 systematically designed queries distributed across 6 distinct categories of varying interaction complexity; and the introduction of MCP-Eval, a novel outcome-oriented evaluation methodology prioritizing real-world task success. Through extensive empirical evaluation of leading language agents, we provide foundational insights. MCP-AgentBench aims to equip the research community with a standardized and reliable framework to build, validate, and advance agents capable of fully leveraging MCP's transformative benefits, thereby accelerating progress toward truly capable and interoperable AI systems.

PDF Details DOI

TAAS Journal 2026 Journal Article

OptimML: Joint Control of Inference Latency and Server Power Consumption for ML Performance Optimization

Guoyu Chen
Xiaorui Wang

Power capping is an important technique for high-density servers to safely oversubscribe the power infrastructure in a data center. However, power capping is commonly accomplished by dynamically lowering the server processors’ frequency levels, which can result in degraded application performance. For servers that run important machine learning (ML) applications with Service-Level Objective (SLO) requirements, inference performance such as recognition accuracy must be optimized within a certain latency constraint, which demands high server performance. To achieve the best inference accuracy under the desired latency and server power constraints, this article proposes OptimML, a multi-input-multi-output (MIMO) control framework that jointly controls both inference latency and server power consumption, by flexibly adjusting the ML model size (and so its required computing resources) when server frequency needs to be lowered for power capping. Our results on a hardware testbed with widely adopted ML framework (including PyTorch, TensorFlow, and MXNet) show that OptimML achieves higher inference accuracy compared with several well-designed baselines, while respecting both latency and power constraints. Furthermore, an adaptive control scheme with online model switching and estimation is designed to achieve analytic assurance of control accuracy and system stability, even in the face of significant workload or hardware variations.

Details DOI

AAAI Conference 2026 Conference Paper

Thinker: Training LLMs in Hierarchical Thinking for Deep Search via Multi-Turn Interaction

Jun Xu
Xinkai Du
Yu Ao
Peilong Zhao
Yang Li
Ling Zhong
Lin Yuan
Zhongpu Bo

Efficient retrieval of external knowledge bases and web pages is crucial for enhancing the reasoning abilities of LLMs. Previous works on training LLMs to leverage external retrievers for solving complex problems have predominantly employed end-to-end reinforcement learning. However, these approaches neglect supervision over the reasoning process, making it difficult to guarantee logical coherence and rigor. To address these limitations, we propose Thinker, a hierarchical thinking model for deep search through multi-turn interaction, making the reasoning process supervisable and verifiable. It decomposes complex problems into independently solvable sub-problems, each dually represented in both natural language and an equivalent logical function to support knowledge base and web searches. Concurrently, dependencies between sub-problems are passed as parameters via these logical functions, enhancing the logical coherence of the problem-solving process. To avoid unnecessary external searches, we perform knowledge boundary determination to check if a sub-problem is within the LLM's intrinsic knowledge, allowing it to answer directly. Experimental results indicate that with as few as several hundred training samples, the performance of Thinker is competitive with established baselines. Furthermore, when scaled to the full training set, Thinker significantly outperforms these methods across various datasets and model sizes.

PDF Details DOI

NeurIPS Conference 2025 Conference Paper

GRIP: A Graph-Based Reasoning Instruction Producer

Jiankang Wang
Jianjun Xu
Xiaorui Wang
Yuxin Wang
Mengting Xing
Shancheng Fang
Hongtao Xie

Large-scale, high-quality data is essential for advancing the reasoning capabilities of large language models (LLMs). As publicly available Internet data becomes increasingly scarce, synthetic data has emerged as a crucial research direction. However, existing data synthesis methods often suffer from limited scalability, insufficient sample diversity, and a tendency to overfit to seed data, which constrains their practical utility. In this paper, we present \textit{\textbf{GRIP}}, a \textbf{G}raph-based \textbf{R}easoning \textbf{I}nstruction \textbf{P}roducer that efficiently synthesizes high-quality and diverse reasoning instructions. \textit{GRIP} constructs a knowledge graph by extracting high-level concepts from seed data, and uniquely leverages both explicit and implicit relationships within the graph to drive large-scale and diverse instruction data synthesis, while employing open-source multi-model supervision to ensure data quality. We apply \textit{GRIP} to the critical and challenging domain of mathematical reasoning. Starting from a seed set of 7. 5K math reasoning samples, we construct \textbf{GRIP-MATH}, a dataset containing 2. 1 million synthesized question-answer pairs. Compared to similar synthetic data methods, \textit{GRIP} achieves greater scalability and diversity while also significantly reducing costs. On mathematical reasoning benchmarks, models trained with GRIP-MATH demonstrate substantial improvements over their base models and significantly outperform previous data synthesis methods.

PDF Details

IJCAI Conference 2025 Conference Paper

IterMeme: Expert-Guided Multimodal LLM for Interactive Meme Creation with Layout-Aware Generation

Yaqi Cai
Shancheng Fang
Yadong Qu
Xiaorui Wang
Meng Shao
Hongtao Xie

Meme creation is a creative process that blends images and text. However, existing methods lack critical components, failing to support intent-driven caption-layout generation and personalized generation, making it difficult to generate high-quality memes. To address this limitation, we propose IterMeme, an end-to-end interactive meme creation framework that utilizes a unified Multimodal Large Language Model (MLLM) to facilitate seamless collaboration among multiple components. To overcome the absence of a caption-layout generation component, we develop a robust layout representation method and construct a large-scale image-caption-layout dataset, MemeCap, which enhances the model’s ability to comprehend emotions and coordinate caption-layout generation effectively. To address the lack of a personalization component, we introduce a parameter-shared dual-LLM architecture that decouples the intricate representations of reference images and text. Furthermore, we incorporate the expert-guided M³OE for fine-grained identity properties (IP) feature extraction and cross-modal fusion. By dynamically injecting features into every layer of the model, we enable adaptive refinement of both visual and semantic information. Experimental results demonstrate that IterMeme significantly advances the field of meme creation by delivering consistently high-quality outcomes. The code, model, and dataset will be open-sourced to the community.

PDF Details DOI

IJCAI Conference 2025 Conference Paper

MIRROR: Multi-agent Intra- and Inter-Reflection for Optimized Reasoning in Tool Learning

Zikang Guo
Benfeng Xu
Xiaorui Wang
Zhendong Mao

Complex tasks involving tool integration pose significant challenges for Large Language Models (LLMs), leading to the emergence of multi-agent workflows as a promising solution. Reflection has emerged as an effective strategy for correcting erroneous trajectories in agentic workflows. However, existing approaches only exploit such capability in the post-action stage, where the agent observes the execution outcomes. We argue that, like humans, LLMs can also engage in reflection before action execution: the agent can anticipate undesirable outcomes from its own decisions, which not only provides a necessarily complementary perspective to evaluate the decision but also prevents the propagation of errors throughout the trajectory. In this paper, we propose MIRROR, a framework that consists of both intra-reflection, which critically assesses intended actions before execution, and inter-reflection, which further adjusts the trajectory based on observations. This design systematically leverages LLM reflection capabilities to eliminate and rectify erroneous actions on a more comprehensive scope. Evaluations on both the StableToolBench and TravelPlanner benchmarks demonstrate MIRROR's superior performance, achieving state-of-the-art results compared to existing approaches.

PDF Details DOI

ICLR Conference 2025 Conference Paper

SAGEPhos: Sage Bio-Coupled and Augmented Fusion for Phosphorylation Site Detection

Jingjie Zhang
Hanqun Cao
Zijun Gao
Xiaorui Wang
Chunbin Gu

Phosphorylation site prediction based on kinase-substrate interaction plays a vital role in understanding cellular signaling pathways and disease mechanisms. Computational methods for this task can be categorized into kinase-family-focused and individual kinase-targeted approaches. Individual kinase-targeted methods have gained prominence for their ability to explore a broader protein space and provide more precise target information for kinase inhibitors. However, most existing individual kinase-based approaches focus solely on sequence inputs, neglecting crucial structural information. To address this limitation, we introduce SAGEPhos (Structure-aware kinAse-substrate bio-coupled and bio-auGmented nEtwork for Phosphorylation site prediction), a novel framework that modifies the semantic space of main protein inputs using auxiliary inputs at two distinct modality levels. At the inter-modality level, SAGEPhos introduces a Bio-Coupled Modal Fusion method, distilling essential kinase sequence information to refine task-oriented local substrate feature space, creating a shared semantic space that captures crucial kinase-substrate interaction patterns. Within the substrate's intra-modality domain, it focuses on Bio-Augmented Fusion, emphasizing 2D local sequence information while selectively incorporating 3D spatial information from predicted structures to complement the sequence space. Moreover, to address the lack of structural information in current datasets, we contribute a new, refined phosphorylation site prediction dataset, which incorporates crucial structural elements and will serve as a new benchmark for the field. Experimental results demonstrate that SAGEPhos significantly outperforms baseline methods, notably achieving almost 10\% and 12\% improvements in prediction accuracy and AUC-ROC, respectively. We further demonstrate our algorithm's robustness and generalization through stable results across varied data partitions and significant improvements in zero-shot scenarios. These results underscore the effectiveness of constructing a larger and more precise protein space in advancing the state-of-the-art in phosphorylation site prediction. We release the SAGEPhos models and code at https://github.com/ZhangJJ26/SAGEPhos.

Details

TAAS Journal 2018 Journal Article

SOD

Marco Brocanelli
Xiaorui Wang

A major concern for today’s smartphones is their much faster battery drain than traditional feature phones, despite their greater battery capacities. The difference is mainly contributed by those more powerful but also much more power-consuming smartphone components, such as the multi-core application processor and the high-definition (HD) display. While the application processor must be active when any smart apps are being used, it is also unnecessarily waken up, even during idle periods, to perform operations related to basic phone functions (i.e., incoming calls and text messages). In addition, the power-hungry HD display is also used unnecessarily for such basic functions. In this article, we investigate how to increase the battery life of smartphones by minimizing the use of application processor and HD display for operations related to basic functions. We find that the application processor is often waken up by a process running on it, called the Radio Interface Layer Daemon (RILD), which interfaces the user and apps to the GSM/LTE cellular network. In particular, we demonstrate that a great amount of energy could be saved if RILD is stopped, such that the application processor can sleep more often. Based on this key finding, we design a Smart On Demand (SOD) configuration that reduces the smartphone energy consumption by running RILD operations on a secondary low-power microcontroller and by using a secondary low-power display to interface the user with basic functions. As a result, basic phone functions can be handled at much lower energy costs and the power-consuming components, i.e., application processor and HD display, are waken up only when one needs to use any smart apps, in an on-demand manner. We have built a hardware prototype of SOD and evaluated it with real user traces. Our results show that SOD can increase its battery life by up to 2.5 more days.

Details DOI