Hallucination in world models is a data coverage issue predictable by three signals and preventable through targeted training sampling and online data collection.
hub
Nitrogen: An open foundation model for generalist gaming agents
15 Pith papers cite this work. Polarity classification is still indexing.
15
Pith papers citing it
hub tools
citation-role summary
background 3
citation-polarity summary
years
2026 15roles
background 3polarities
background 3representative citing papers
Equation-to-Behavior Prompting lets large LLMs match cognitive models like Bayesian updating in persuasion games; RL training cuts small-model belief error by 26.5% and improves diverse training outcomes by 2.5-12%.
OmniGameArena is a unified UE5 benchmark with 12 games and the IDC harness for cold-start scores and improvement dynamics of VLM agents.
Prisma-World is a diffusion-based multi-agent video model that uses joint full-attention, multi-agent RoPE, and relative camera geometry injection plus curriculum training to produce consistent cross-view videos from flexible agent counts.
SCOPE adds per-pixel action conditioning to pretrained video diffusion models and releases the CrossFPS multi-game dataset to support cross-game FPS world model simulation with zero-shot transfer.
SPIKE dual-controller framework raises success rates 5-9 points and cuts tokens 55% in StarDojo agents by reusing strategic plans across stable segments and escalating only at detected events.
Fine-tuning VLMs to output action sequences for puzzles causes emergent internal visual representations that improve performance when integrated into reasoning.
Odysseus adapts PPO with a turn-level critic and leverages pretrained VLM action priors to train agents achieving at least 3x average game progress over frontier models in long-horizon Super Mario Land.
The manuscript body introduces PokeGym, a vision-only automated 3D-game benchmark, while the abstract claims a G-EvoMAC method and 60.18% success rate absent from the body.
Generalizable agents require environment scaling via diverse executable rule-sets, distinguished from trajectory and task scaling in a new taxonomy.
GROW decomposes trajectories into state-action samples to enable GRPO for multi-turn VLM agents and reports state-of-the-art results on more than 800 Minecraft tasks.
The paper organizes research on generalist game AI into Dataset, Model, Harness, and Benchmark pillars and charts a five-level progression from single-game mastery to agents that create and live inside game multiverses.
Proposes a requirements-based framework for RL-augmented game AI, discusses deployment practicalities, and identifies research bottlenecks for industry adoption.
A survey proposing a three-level capability taxonomy (L1 Predictor, L2 Simulator, L3 Evolver) for world models across physical, digital, social, and scientific domains.
A preview system demonstrates real-time controllable world modeling at 14-15 FPS on RTX 4090 by adapting open video backbones with action pathways for keyboard/mouse control and multimodal features.
citing papers explorer
-
Hallucination in World Models is Predictable and Preventable
Hallucination in world models is a data coverage issue predictable by three signals and preventable through targeted training sampling and online data collection.
-
Using Cognitive Models to Improve Language Model Simulation of Human Persuasion Games
Equation-to-Behavior Prompting lets large LLMs match cognitive models like Bayesian updating in persuasion games; RL training cuts small-model belief error by 26.5% and improves diverse training outcomes by 2.5-12%.
-
OmniGameArena: A Unified UE5 Benchmark for VLM Game Agents with Improvement Dynamics
OmniGameArena is a unified UE5 benchmark with 12 games and the IDC harness for cold-start scores and improvement dynamics of VLM agents.
-
Prisma-World: Camera-Controllable Multi-Agent Video World Model
Prisma-World is a diffusion-based multi-agent video model that uses joint full-attention, multi-agent RoPE, and relative camera geometry injection plus curriculum training to produce consistent cross-view videos from flexible agent counts.
-
SCOPE: Simulating Cross-game Operations in Playable Environments for FPS World Models
SCOPE adds per-pixel action conditioning to pretrained video diffusion models and releases the CrossFPS multi-game dataset to support cross-game FPS world model simulation with zero-shot transfer.
-
SPIKE: An Adaptive Dual Controller Framework for Cost-Efficient Long-Horizon Game Agents
SPIKE dual-controller framework raises success rates 5-9 points and cuts tokens 55% in StarDojo agents by reusing strategic plans across stable segments and escalating only at detected events.
-
Do multimodal models imagine electric sheep?
Fine-tuning VLMs to output action sequences for puzzles causes emergent internal visual representations that improve performance when integrated into reasoning.
-
Odysseus: Scaling VLMs to 100+ Turn Decision-Making in Games via Reinforcement Learning
Odysseus adapts PPO with a turn-level critic and leverages pretrained VLM action priors to train agents achieving at least 3x average game progress over frontier models in long-horizon Super Mario Land.
-
Mastering PokeGym: Graph-Guided Multimodal Evolution at Test Time
The manuscript body introduces PokeGym, a vision-only automated 3D-game benchmark, while the abstract claims a G-EvoMAC method and 60.18% success rate absent from the body.
-
Scalable Environments Drive Generalizable Agents
Generalizable agents require environment scaling via diverse executable rule-sets, distinguished from trajectory and task scaling in a new taxonomy.
-
GROW: Aligning GRPO with State-Action Modeling for Open-World VLM Agents
GROW decomposes trajectories into state-action samples to enable GRPO for multi-turn VLM agents and reports state-of-the-art results on more than 800 Minecraft tasks.
-
Towards Generalist Game Players: An Investigation of Foundation Models in the Game Multiverse
The paper organizes research on generalist game AI into Dataset, Model, Harness, and Benchmark pillars and charts a five-level progression from single-game mastery to agents that create and live inside game multiverses.
-
Augmenting Game AI with Deep Reinforcement Learning
Proposes a requirements-based framework for RL-augmented game AI, discusses deployment practicalities, and identifies research bottlenecks for industry adoption.
-
Agentic World Modeling: Foundations, Capabilities, Laws, and Beyond
A survey proposing a three-level capability taxonomy (L1 Predictor, L2 Simulator, L3 Evolver) for world models across physical, digital, social, and scientific domains.
-
DreamForge-World 0.1 Preview: A Low-Compute Real-Time Controllable World Model
A preview system demonstrates real-time controllable world modeling at 14-15 FPS on RTX 4090 by adapting open video backbones with action pathways for keyboard/mouse control and multimodal features.