arXiv subjects

Michael Choi

Publications and source records attributed to Michael Choi.

4 recordsLinked to original sources

arXiv2025

Iterative Multi-Agent Reinforcement Learning: A Novel Approach Toward Real-World Multi-Echelon Inventory Optimization

Multi-echelon inventory optimization (MEIO) is critical for effective supply chain management, but its inherent complexity can pose significant challenges. Heuristics are commonly used to address this complexity, yet they often face limitations in scope and scalability. Recent research has found deep reinforcement learning (DRL) to be a promising alternative to traditional heuristics, offering greater versatility by utilizing dynamic decision-making capabilities. However, since DRL is known to struggle with the curse of dimensionality, its relevance to complex real-life supply chain scenarios is still to be determined. This thesis investigates DRL's applicability to MEIO problems of increasing complexity. A state-of-the-art DRL model was replicated, enhanced, and tested across 13 supply chain scenarios, combining diverse network structures and parameters. To address DRL's challenges with dimensionality, additional models leveraging graph neural networks (GNNs) and multi-agent reinforcement learning (MARL) were developed, culminating in the novel iterative multi-agent reinforcement learning (IMARL) approach. IMARL demonstrated superior scalability, effectiveness, and reliability in optimizing inventory policies, consistently outperforming benchmarks. These findings confirm the potential of DRL, particularly IMARL, to address real-world supply chain challenges and call for additional research to further expand its applicability.

cs.LG↗

arXiv2025

Jailbreaking to Jailbreak

Large Language Models (LLMs) can be used to red team other models (e.g. jailbreaking) to elicit harmful contents. While prior works commonly employ open-weight models or private uncensored models for doing jailbreaking, as the refusal-training of strong LLMs (e.g. OpenAI o3) refuse to help jailbreaking, our work turn (almost) any black-box LLMs into attackers. The resulting $J_2$ (jailbreaking-to-jailbreak) attackers can effectively jailbreak the safeguard of target models using various strategies, both created by themselves or from expert human red teamers. In doing so, we show their strong but under-researched jailbreaking capabilities. Our experiments demonstrate that 1) prompts used to create $J_2$ attackers transfer across almost all black-box models; 2) an $J_2$ attacker can jailbreak a copy of itself, and this vulnerability develops rapidly over the past 12 months; 3) reasong models, such as Sonnet-3.7, are strong $J_2$ attackers compared to others. For example, when used against the safeguard of GPT-4o, $J_2$ (Sonnet-3.7) achieves 0.975 attack success rate (ASR), which matches expert human red teamers and surpasses the state-of-the-art algorithm-based attacks. Among $J_2$ attackers, $J_2$ (o3) achieves highest ASR (0.605) against Sonnet-3.5, one of the most robust models.

cs.CL↗

arXiv2025

Humanity's Last Exam

Benchmarks are important tools for tracking the rapid advancements in large language model (LLM) capabilities. However, benchmarks are not keeping pace in difficulty: LLMs now achieve over 90\% accuracy on popular benchmarks like MMLU, limiting informed measurement of state-of-the-art LLM capabilities. In response, we introduce Humanity's Last Exam (HLE), a multi-modal benchmark at the frontier of human knowledge, designed to be the final closed-ended academic benchmark of its kind with broad subject coverage. HLE consists of 2,500 questions across dozens of subjects, including mathematics, humanities, and the natural sciences. HLE is developed globally by subject-matter experts and consists of multiple-choice and short-answer questions suitable for automated grading. Each question has a known solution that is unambiguous and easily verifiable, but cannot be quickly answered via internet retrieval. State-of-the-art LLMs demonstrate low accuracy and calibration on HLE, highlighting a significant gap between current LLM capabilities and the expert human frontier on closed-ended academic questions. To inform research and policymaking upon a clear understanding of model capabilities, we publicly release HLE at https://lastexam.ai.

cs.LG↗

arXiv2018

SOFIA - HIRMES: Looking forward to the HIgh-Resolution Mid-infrarEd Spectrometer

The HIgh-Resolution Mid-infrarEd Spectrometer (HIRMES) is the 3rd Generation Instrument for the Stratospheric Observatory For Infrared Astronomy (SOFIA), currently in development at the NASA Goddard Space Flight Center (GSFC), and due for commissioning in 2019. By combining direct-detection Transition Edge Sensor (TES) bolometer arrays, grating-dispersive spectroscopy, and a host of Fabry-Perot tunable filters, HIRMES will provide the ability for High Resolution (R~100,000), Mid-Resolution (R~10,000), and Low-Resolution (R~600) slit-spectroscopy, and 2D Spectral Imaging (R~2000 at selected wavelengths) over the 25 - 122 μm mid-far infrared waveband. The driving science application is the evolution of proto-planetary systems via measurements of water-vapor, water-ice, deuterated hydrogen (HD), and neutral oxygen lines. However, HIRMES has been designed to be as flexible as possible to cover a wide range of science cases that fall within its phase-space, all whilst reaching sensitivities and observing powers not yet seen thus far on SOFIA, providing unique observing capabilities which will remain unmatched for decades.

astro-ph.IM↗