arXiv subjects

Heming Fu

Publications and source records attributed to Heming Fu.

7 recordsLinked to original sources

arXiv2026

Not All Tokens Are Equal: Inflation-Aware Routing for Agentic LLM Systems

When a language model fails to answer a query on the first attempt, an agentic system retries, consuming additional tokens each time. This retry overhead creates a gap between what a model's per-token price implies and what a full workflow actually costs. We call this gap \emph{token inflation} and define it as the ratio of true workflow cost to single-call cost. Systems like FrugalGPT route based on the latter, which can underestimate real cost by more than $2\times$ on difficult tasks. We address this with InflationAgent, a four-stage router that (1) measures token inflation systematically across model tiers and task types, finding inflation as high as $4.25\times$ for a 7B model on multi-hop question answering; (2) introduces CoT Branching Entropy (CBE), a pre-execution difficulty signal computed entirely from local inference, which predicts high inflation with AUROC 0.887; and (3) selects models by maximizing a Semantic Exchange Rate (SER) that divides expected accuracy by predicted true cost, with a fresh-escalation policy that discards failed chains before routing to a stronger model. On GSM8K under a fixed budget, InflationAgent achieves 94.7\% accuracy versus 91.0\% for FrugalGPT while using 31\% fewer tokens, and we show that forwarding a failed reasoning chain to GPT-4o reduces its accuracy by up to 34.8 percentage points, validating the fresh-escalation design.

cs.CL↗

arXiv2025

MARLIN: Multi-Agent Reinforcement Learning with Murmuration Intelligence and LLM Guidance for Reservoir Management

As climate change intensifies extreme weather events, water disasters pose growing threats to global communities, making adaptive reservoir management critical for protecting vulnerable populations and ensuring water security. Modern water resource management faces unprecedented challenges from cascading uncertainties propagating through interconnected reservoir networks. These uncertainties, rooted in physical water transfer losses and environmental variability, make precise control difficult. For example, sending 10 tons downstream may yield only 8-12 tons due to evaporation and seepage. Traditional centralized optimization approaches suffer from exponential computational complexity and cannot effectively handle such real-world uncertainties, while existing multi-agent reinforcement learning (MARL) methods fail to achieve effective coordination under uncertainty. To address these challenges, we present MARLIN, a decentralized reservoir management framework inspired by starling murmurations intelligence. Integrating bio-inspired alignment, separation, and cohesion rules with MARL, MARLIN enables individual reservoirs to make local decisions while achieving emergent global coordination. In addition, a LLM provides real-time reward shaping signals, guiding agents to adapt to environmental changes and human-defined preferences. Experiments on USGS data show that MARLIN improves uncertainty handling by 23\%, cuts computation by 35\%, and accelerates flood response by 68\%, exhibiting super-linear coordination, with complexity scaling 5.4x from 400 to 10,000 nodes. These results demonstrate MARLIN's potential for disaster prevention and protecting communities through intelligent, scalable water resource management.

cs.MA↗

arXiv2025

MultiFinBen: Benchmarking Large Language Models for Multilingual and Multimodal Financial Application

Real-world financial analysis involves information across multiple languages and modalities, from reports and news to scanned filings and meeting recordings. Yet most existing evaluations of LLMs in finance remain text-only, monolingual, and largely saturated by current models. To bridge these gaps, we present MultiFinBen, the first expert-annotated multilingual (five languages) and multimodal (text, vision, audio) benchmark for evaluating LLMs in realistic financial contexts. MultiFinBen introduces two new task families: multilingual financial reasoning, which tests cross-lingual evidence integration from filings and news, and financial OCR, which extracts structured text from scanned documents containing tables and charts. Rather than aggregating all available datasets, we apply a structured, difficulty-aware selection based on advanced model performance, ensuring balanced challenge and removing redundant tasks. Evaluating 21 leading LLMs shows that even frontier multimodal models like GPT-4o achieve only 46.01% overall, stronger on vision and audio but dropping sharply in multilingual settings. These findings expose persistent limitations in multilingual, multimodal, and expert-level financial reasoning. All datasets, evaluation scripts, and leaderboards are publicly released.

cs.CL↗

arXiv2025

Multi-Agent Reinforcement Learning for Decentralized Reservoir Management via Murmuration Intelligence

Conventional centralized water management systems face critical limitations from computational complexity and uncertainty propagation. We present MurmuRL, a novel decentralized framework inspired by starling murmurations intelligence, integrating bio-inspired alignment, separation, and cohesion rules with multi-agent reinforcement learning. MurmuRL enables individual reservoirs to make autonomous local decisions while achieving emergent global coordination. Experiments on grid networks demonstrate that MurmuRL achieves 8.8% higher final performance while using 27% less computing overhead compared to centralized approaches. Notably, strategic diversity scales super-linearly with system size, exhibiting sophisticated coordination patterns and enhanced resilience during extreme events. MurmuRL offers a scalable solution for managing complex water systems by leveraging principles of natural collective behavior.

eess.SY↗

arXiv2025

SHADE-AD: An LLM-Based Framework for Synthesizing Activity Data of Alzheimer's Patients

Alzheimer's Disease (AD) has become an increasingly critical global health concern, which necessitates effective monitoring solutions in smart health applications. However, the development of such solutions is significantly hindered by the scarcity of AD-specific activity datasets. To address this challenge, we propose SHADE-AD, a Large Language Model (LLM) framework for Synthesizing Human Activity Datasets Embedded with AD features. Leveraging both public datasets and our own collected data from 99 AD patients, SHADE-AD synthesizes human activity videos that specifically represent AD-related behaviors. By employing a three-stage training mechanism, it broadens the range of activities beyond those collected from limited deployment settings. We conducted comprehensive evaluations of the generated dataset, demonstrating significant improvements in downstream tasks such as Human Activity Recognition (HAR) detection, with enhancements of up to 79.69%. Detailed motion metrics between real and synthetic data show strong alignment, validating the realism and utility of the synthesized dataset. These results underscore SHADE-AD's potential to advance smart health applications by providing a cost-effective, privacy-preserving solution for AD monitoring.

cs.LG↗

arXiv2024

Visualizing the Invisible: A Generative AR System for Intuitive Multi-Modal Sensor Data Presentation

Understanding sensor data can be difficult for non-experts because of the complexity and different semantic meanings of sensor modalities. This leads to a need for intuitive and effective methods to present sensor information. However, creating intuitive sensor data visualizations presents three key challenges: the variability of sensor readings, gaps in domain comprehension, and the dynamic nature of sensor data. To address these issues, we propose Vivar, a novel system that integrates multi-modal sensor data and presents 3D volumetric content for AR visualization. In particular, we introduce a cross-modal embedding approach that maps sensor data into a pre-trained visual embedding space through barycentric interpolation. This approach accurately reflects value changes in multi-modal sensor information, ensuring that sensor variations are properly shown in visualization outcomes. Vivar also incorporates sensor-aware AR scene generation using foundation models and 3D Gaussian Splatting (3DGS) without requiring domain expertise. In addition, Vivar leverages latent reuse and caching strategies to accelerate 2D and AR content generation, demonstrating 11x latency reduction without compromising quality. A user study involving over 503 participants, including domain experts, demonstrates Vivar's effectiveness in accuracy, consistency, and real-world applicability, paving the way for more intuitive sensor data visualization.

cs.HC↗

arXiv2023

ADMarker: A Multi-Modal Federated Learning System for Monitoring Digital Biomarkers of Alzheimer's Disease

Alzheimer's Disease (AD) and related dementia are a growing global health challenge due to the aging population. In this paper, we present ADMarker, the first end-to-end system that integrates multi-modal sensors and new federated learning algorithms for detecting multidimensional AD digital biomarkers in natural living environments. ADMarker features a novel three-stage multi-modal federated learning architecture that can accurately detect digital biomarkers in a privacy-preserving manner. Our approach collectively addresses several major real-world challenges, such as limited data labels, data heterogeneity, and limited computing resources. We built a compact multi-modality hardware system and deployed it in a four-week clinical trial involving 91 elderly participants. The results indicate that ADMarker can accurately detect a comprehensive set of digital biomarkers with up to 93.8% accuracy and identify early AD with an average of 88.9% accuracy. ADMarker offers a new platform that can allow AD clinicians to characterize and track the complex correlation between multidimensional interpretable digital biomarkers, demographic factors of patients, and AD diagnosis in a longitudinal manner.

cs.LG↗