1 citations · 2 across the 11 of their papers we have counts for
22 papers
Tandem: Serving Hybrid Linear-Attention GUI Agents on Mobile NPUs
Zhuomin Chen, Jingchao Ni, Xu Zheng +4
Smartphone GUI agents complete tasks in mobile apps by repeatedly reading the screen and asking a model for the next action. Mobile neural processing units (NPUs) enable local exec…
First-Order LoRA Training with Phone NPUs
Xu Zheng, Wei Cheng, Zhuomin Chen +4
On-device fine-tuning lets a large language model (LLM) learn from its user's private data without uploading it. Low-rank adaptation (LoRA) reduces the trainable parameters, but ea…
Muse Glimmer 30B: An On-Device Agentic Model
Arash Ebrahim Abtahizadeh, Muna Aghamelu, Adil Ahmed +327
We introduce Muse Glimmer, an open-source 30B parameter causal language model with perception encoder, purpose-built for autonomous agentic tasks on consumer hardware. Muse Glimmer…
SpecForge: Recursive Self-Improvement for Coding Agents via Behavioral Specifications
Yutong Cheng, Haifeng Chen, Peng Gao +1
Coding agents based on frozen language models can improve their performance by refining the external context they rely on, without updating model weights. We study recursive self-i…
WorkflowOps: Learning Agent Collaboration Priors for Multi-Agent Workflow Orchestration
Qi Cheng, Shengyu Chen, Wei Cheng +4
Multi-agent systems are increasingly deployed for complex knowledge work, yet their orchestration layers remain largely memoryless: each new task is decomposed, assigned, and execu…
RA-MoWE: Workflow-Affinity Embeddings for Query Clustering and Agentic Workflow Generation
Qi Cheng, Shengyu Chen, Wei Cheng +4
Agentic workflows enable large language models (LLMs) to solve complex tasks by coordinating reasoning, tool use, and verification. However, a workflow optimized for an entire task…
OOPMAS: Object-Oriented Multi-Agent Systems for Query-Level Workflow Generation
Qi Cheng, Shengyu Chen, Wei Cheng +4
Multi-agent systems (MAS) powered by large language models have shown strong performance across code generation, mathematical reasoning, and question answering. However, existing m…
ST-Bench: A Spatial-Temporal Benchmark for Multi-Agent System Generation on Scientific Research Tasks
Qi Cheng, Rongchao Dong, Shengyu Chen +8
The rapid progress of LLM-based multi-agent systems (MAS) has shown that they largely outperform single agents on coding, math, and QA tasks, where executable tests provide a binar…
Do Time-Series QA Systems Read the Time Series? Evidence Use and Reasoning Reliability
Zhuomin Chen, Jingchao Ni, Xu Zheng +4
In recent years, time-series question answering (QA) systems have made significant progress. However, generating a correct answer does not show whether retaining the supplied numer…
Pay for the Fault, Not the Flow: Label-Free In-Flow Multi-Agent Workflow Optimization
Xuehang Guo, Haoyu Wang, Shengyu Chen +4
Large language models (LLMs) increasingly construct multi-agent workflows that decompose a complex task and assign specialist agents from a pool. However, building such a workflow…
MetaCaster: Meta-Harness-Optimized Agent for End-to-End Few-Shot Learning of Lightweight Time Series Forecasters
ChengAo Shen, Wenchao Yu, Fangyu Wu +6
Time series forecasting (TSF) is evolving toward multimodal and agentic settings, yet using foundation models remains uneconomical in resource-constrained scenarios, where compact,…
Towards A Unified Information Bottleneck Framework for Time Series Explanations
Xu Zheng, Zichuan Liu, Zhuomin Chen +7
Explaining deep learning models operating on time series data is crucial in various applications that require transparent and interpretable insights into model behavior. {Existing…
Information Bottleneck Learning for Faithful Time Series Forecasting Explanations
Xu Zheng, Wei Cheng, Zhuomin Chen +3
As forecasts increasingly drive decisions in fields such as energy, transportation, and healthcare, understanding the historical data behind these predictions has become as crucial…
Leveraging Trajectory Graphs for Pre-Execution Error Diagnosis in Agentic LLM Systems
Xu Zheng, Zhuomin Chen, Chaohao Lin +4
Large Language Model~(LLM)-based agents have demonstrated exceptional performance across a wide range of complex interactive tasks. However, they often struggle with long-horizon i…
Bridging Inference-Time Scaling and Episodic Memory with Action-Centric Graphs
Xu Zheng, Chaohao Lin, Zhuomin Chen +4
Recent advancements in inference-time scaling have significantly unlocked the complex reasoning capabilities of Large Language Models~(LLMs). However, for agents, these approaches…
DecoSearch: Complexity-Aware Routing and Plan-Level Repair for Text-to-SQL
Esteban Schafir, Xu Zheng, Hojat Allah Salehi +4
Large Language Models (LLMs) have demonstrated remarkable capabilities in translating natural language to SQL, yet existing methods still falter on complex queries requiring multi-…
Uncovering Insights of Compound Flooding with Data-Driven AI
Xu Zheng, Chaohao Lin, Sipeng Chen +7
Compound flooding, driven by nonlinear interactions between multiple hydrometeorological factors, poses a significant challenge to hazard prevention. Existing forecasting approache…
Towards Structurally Explainable Machine-Generated Text Detection: A Graph-Perspective Framework
Xu Zheng, Zhuomin Chen, Esteban Schafir +7
Despite the success of machine-generated text detectors, the black-box nature remains a critical limitation. Traditional explainability methods rely on token-level saliency, insuff…
F-Fidelity: A Robust Framework for Faithfulness Evaluation of Explainable AI
Xu Zheng, Farhad Shirani, Zhuomin Chen +4
Recent research has developed a number of eXplainable AI (XAI) techniques, such as gradient-based approaches, input perturbation-base methods, and black-box explanation methods. Wh…
Protecting Your LLMs with Information Bottleneck
Zichuan Liu, Zefan Wang, Linjie Xu +6
The advent of large language models (LLMs) has revolutionized the field of natural language processing, yet they might be attacked to produce harmful content. Despite efforts to et…
Parametric Augmentation for Time Series Contrastive Learning
Xu Zheng, Tianchun Wang, Wei Cheng +4
Modern techniques like contrastive learning have been effectively used in many areas, including computer vision, natural language processing, and graph-structured data. Creating po…
PAC Learnability under Explanation-Preserving Graph Perturbations
Xu Zheng, Farhad Shirani, Tianchun Wang +4
Graphical models capture relations between entities in a wide range of applications including social networks, biology, and natural language processing, among others. Graph neural…