Paul Richmond

Paul Richmond, Prarit Agarwal, Borun Chowdhury et al.

We present specialized Large Language Models for theoretical High-Energy Physics, obtained as 20 fine-tuned variants of the 8-billion parameter Llama-3.1 model. Each variant was trained on arXiv abstracts (through August 2024) from different combinations of hep-th, hep-ph and gr-qc. For a comparative study, we also trained models on datasets that contained abstracts from disparate fields such as the q-bio and cs categories. All models were fine-tuned using two distinct Low-Rank Adaptation fine-tuning approaches and varying dataset sizes, and outperformed the base model on hep-th abstract completion tasks. We compare performance against leading commercial LLMs (ChatGPT, Claude, Gemini, DeepSeek) and derive insights for further developing specialized language models for High-Energy Theoretical Physics.

John Charlton, Luis Rene Montana Gonzalez, Steve Maddock et al.

Understanding how people view and interact with autonomous vehicles is important to guide future directions of research. One such way of aiding understanding is through simulations of virtual environments involving people and autonomous vehicles. We present a simulation model that incorporates people and autonomous vehicles in a shared urban space. The model is able to simulate many thousands of people and vehicles in real-time. This is achieved by use of GPU hardware, and through a novel linear program solver optimized for large numbers of problems on the GPU. The model is up to 30 times faster than the equivalent multi-core CPU model.

John Charlton, Luis Rene Montana Gonzalez, Steve Maddock et al.

Real-time large-scale crowd simulations with realistic behavior, are important for many application areas. On CPUs, the ORCA pedestrian steering model is often used for agent-based pedestrian simulations. This paper introduces a technique for running the ORCA pedestrian steering model on the GPU. Performance improvements of up to 30 times greater than a multi-core CPU model are demonstrated. This improvement is achieved through a specialized linear program solver on the GPU and spatial partitioning of information sharing. This allows over 100,000 people to be simulated in real time (60 frames per second).