A

adversarial

Empirical validation of C4 geometric defense against 16 Agents of Chaos. 550 adversarial prompts. 4 defense systems. 96.7% block rate. LLM validation on GPT-4o-mini + Mistral 7B. MIT.

A minimal Python3 program to collect and store JSON responses concurrently from LLM models using zai-sdk.

The repository contains an Unreal Engine project for a strategy role playing videogame. The game's combat complexity is low-medium as of the current implementation. The project is meant to be used to test search based adversarial algorithms.

Implemented: Alpha-beta, Negascout, MTD(f), MonteCarloTreeSearch(UCT), MCTS(Epsilon-greedy) and RAVE. There is an incomplete version of UCT-RAVE.

An implementation of a Monte Carlo tree search adversarial agent