arXiv subjects

Mohamad Hagog

Publications and source records attributed to Mohamad Hagog.

3 recordsLinked to original sources

arXiv2025

Regress, Don't Guess -- A Regression-like Loss on Number Tokens for Language Models

While language models have exceptional capabilities at text generation, they lack a natural inductive bias for emitting numbers and thus struggle in tasks involving quantitative reasoning, especially arithmetic. One fundamental limitation is the nature of the cross-entropy (CE) loss, which assumes a nominal scale and thus cannot convey proximity between generated number tokens. In response, we here present a regression-like loss that operates purely on token level. Our proposed Number Token Loss (NTL) comes in two flavors and minimizes either the $L_p$ norm or the Wasserstein distance between the numerical values of the real and predicted number tokens. NTL can easily be added to any language model and extend the CE objective during training without runtime overhead. We evaluate the proposed scheme on various mathematical datasets and find that it consistently improves performance in math-related tasks. In a direct comparison on a regression task, we find that NTL can match the performance of a regression head, despite operating on token level. Finally, we scale NTL up to 3B parameter models and observe improved performance, demonstrating its potential for seamless integration into LLMs. We hope to inspire LLM developers to improve their pretraining objectives and distribute NTL as a minimalistic and lightweight PyPI package $ntloss$: https://github.com/ai4sd/number-token-loss. Development code for full paper reproduction is available separately.

cs.CL↗

arXiv2025

First Experience with Real-Time Control Using Simulated VQC-Based Quantum Policies

This paper investigates the integration of quantum computing into offline reinforcement learning and the deployment of the resulting quantum policy in a real-time control hardware realization of the cart-pole system. Variational Quantum Circuits (VQCs) are used to represent the policy. Classical model-based offline policy search was applied, in which a pure VQC with trainable input-output weights is used as a policy network instead of a classical multilayer perceptron. The goal is to evaluate the potential of deploying quantum architectures in real-world industrial control problems. The experimental results show that the investigated model-based offline policy search is able to generate quantum policies that can balance the hardware cart-pole. A latency analysis reveals that while local simulated execution meets real-time requirements, cloud-based quantum processing remains too slow for closed-loop control.

quant-ph↗

arXiv2024

Quantum Advantage Actor-Critic for Reinforcement Learning

Quantum computing offers efficient encapsulation of high-dimensional states. In this work, we propose a novel quantum reinforcement learning approach that combines the Advantage Actor-Critic algorithm with variational quantum circuits by substituting parts of the classical components. This approach addresses reinforcement learning's scalability concerns while maintaining high performance. We empirically test multiple quantum Advantage Actor-Critic configurations with the well known Cart Pole environment to evaluate our approach in control tasks with continuous state spaces. Our results indicate that the hybrid strategy of using either a quantum actor or quantum critic with classical post-processing yields a substantial performance increase compared to pure classical and pure quantum variants with similar parameter counts. They further reveal the limits of current quantum approaches due to the hardware constraints of noisy intermediate-scale quantum computers, suggesting further research to scale hybrid approaches for larger and more complex control tasks.

quant-ph↗