# Terra Nova: A Comprehensive Challenge Environment for Intelligent Agents

arXiv: https://arxiv.org/abs/2511.15378 · HTML: https://mlsanity.com/paper/2511.15378/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Placed in Reasoning for agents, planning and code, one of 17 directions in LLM reasoning.
- Builds on 8 of its 12 references that are in the map.
- First posted on arXiv 19 Nov 2025.

## Placed in

- LLM reasoning / [Reasoning for agents, planning and code](https://mlsanity.com/lr/reasoning-for-agents-planning-and-code/)

## Builds on (most cited first, 8 of 8)

- [Dota 2 with Large Scale Deep Reinforcement Learning](https://mlsanity.com/paper/1912.06680/) (2019)
- [The NetHack Learning Environment](https://mlsanity.com/paper/2006.13760/) (2020)
- [OpenSpiel: A Framework for Reinforcement Learning in Games](https://mlsanity.com/paper/1908.09453/) (2019)
- [Craftax: A Lightning-Fast Benchmark for Open-Ended Reinforcement Learning](https://mlsanity.com/paper/2402.16801/) (2024)
- [Neural MMO: A Massively Multiagent Game Environment for Training and Evaluating Intelligent Agents](https://mlsanity.com/paper/1903.00784/) (2019)
- [CivRealm: A Learning and Reasoning Odyssey in Civilization for Decision-Making Agents](https://mlsanity.com/paper/2401.10568/) (2024)
- [No Press Diplomacy: Modeling Multi-Agent Gameplay](https://mlsanity.com/paper/1909.02128/) (2019)
- [Towards General Negotiation Strategies with End-to-End Reinforcement Learning](https://mlsanity.com/paper/2406.15096/) (2024)

## Abstract (arXiv)

We introduce Terra Nova, a new comprehensive challenge environment (CCE) for reinforcement learning (RL) research inspired by Civilization V. A CCE is a single environment in which multiple canonical RL challenges (e.g., partial observability, credit assignment, representation learning, enormous action spaces, etc.) arise simultaneously. Mastery therefore demands integrated, long-horizon understanding across many interacting variables. We emphasize that this definition excludes challenges that only aggregate unrelated tasks in independent, parallel streams (e.g., learning to play all Atari games at once). These aggregated multitask benchmarks primarily asses whether an agent can catalog and switch among unrelated policies rather than test an agent's ability to perform deep reasoning across many interacting challenges.