arXiv subjects

Jinpeng Miao

Publications and source records attributed to Jinpeng Miao.

4 recordsLinked to original sources

arXiv2026

FinBoardBench: Benchmarking Dynamic Wealth Management and Strategic Financial Reasoning of LLMs via Board Game Simulations

Recently, large language models (LLMs) have achieved superior performance in static financial reasoning and simple dynamic trading tasks. However, existing static financial benchmarks are insufficient to assess the dynamic wealth management and financial decision-making capabilities of LLMs in real-world environments. To bridge this gap, we present FinBoardBench, an evaluation suite based on three classic financial board games: Cashflow, Acquire, and Monopoly. FinBoardBench assesses a comprehensive set of financial skills, including personal cash flow management with debt balancing, corporate investment and acquisition forecasting, and competitive trade negotiations with asset auctions. Our experiments with 9 advanced LLMs reveal that while exhibiting basic long-term planning and investment logic, they fail to effectively leverage complex interactions for profit, and their strong static reasoning performance does not transform into successful dynamic decision-making. Notably, they tend to prioritize immediate asset acquisition over maintaining sufficient liquidity, making them vulnerable to financial crises triggered by random events. We hope that FinBoardBench can provide a valuable reference for more intelligent LLM-based decision-making systems in the future.

cs.CL↗

arXiv2026

Muse Spark Safety & Preparedness Report

Muse Spark is the latest large language model developed by Meta. In this report, we first present evaluations for catastrophic risk domains under Meta's Advanced AI Scaling Framework, along with the evidence that informed our launch decision. We then discuss additional considerations, such as Muse Spark's broader content safety and behavioral profile, that are relevant to overall safety but fall outside the catastrophic risk domains governed by the Framework. Our preparedness results covering Chemical and Biological, Cybersecurity, and Loss of Control risks assess Muse Spark's deployment within Meta AI as presenting acceptable levels of residual risks under our Advanced AI Scaling Framework. We conducted a broad set of evaluations targeting dual-use and high-risk capabilities across these catastrophic risk domains. Those evaluations identified elevated risks prior to mitigations, with Chemical and Biological capabilities assessed as likely reaching the "high risk" category under the Advanced AI Scaling Framework before safeguards were applied. We have implemented a multi-layered set of mitigations that address the identified risks, and Muse Spark demonstrates state-of-the-art refusal across a range of benchmarks related to hazardous workflows in chemistry and biology. We therefore release Muse Spark as the underlying model of Meta AI.

cs.CY↗

arXiv2026

Code World Model Preparedness Report

This report documents the preparedness assessment of Code World Model (CWM), a model for code generation and reasoning about code from Meta. We conducted pre-release testing across domains identified in our Frontier AI Framework as potentially presenting catastrophic risks, and also evaluated the model's misaligned propensities. Our assessment found that CWM does not pose additional frontier risks beyond those present in the current AI ecosystem. We therefore release it as an open-weight model.

cs.SE↗

arXiv2023

A Fog-based Smart Agriculture System to Detect Animal Intrusion

Smart agriculture is one of the most promising areas where IoT-enabled technologies have the potential to substantially improve the quality and quantity of the crops and reduce the associated operational cost. However, building a smart agriculture system presents several challenges, including high latency and bandwidth consumption associated with cloud computing, frequent Internet disconnections in rural areas, and the need to keep costs low for farmers. This paper presents an end-to-end, fog-based smart agriculture infrastructure that incorporates edge computing and LoRa-based communication to address these challenges. Our system is deployed to transform traditional agriculture land of rural areas into smart agriculture. We address the top concern of farmers - animals intruding - by proposing a solution that detects animal intrusion using low-cost PIR sensors, cameras, and computer vision. In particular, we propose three different sensor layouts and a novel algorithm for predicting animals' future locations. Our system can detect animals before they intrude into the field, identify them, predict their future locations, and alert farmers in a timely manner. Our experiments show that the system can effectively and quickly detect animal intrusions while maintaining a much lower cost than current state-of-the-art systems.

eess.SY↗