Duon Labs Logo
01 RESEARCH

World Model Research

Open research powering the world model. Foundations in distribution modeling, scaling laws, and simulation infrastructure.

02 FOUNDATIONS

Research Foundations

Distribution Modeling

SICK BIDE: Binary Implicit Distribution Encoding via GPU-optimized softmax integral computation.

Scaling Laws

Apogée: How world model capabilities scale with compute, data, and model size.

Information Theory

Extropy as predictive information. Signal density, volatility clustering, temporal dependencies.

GPU Infrastructure

Custom GPU kernels, Monte Carlo simulation, model evaluation pipelines.

World Modeling

Learned simulators capturing market dynamics: regime transitions, tail behavior, multi-horizon rollouts.

03 PAPERS

Research Papers

SICK BIDE: Softmax Integral Compute Kernel + Binary Implicit Distribution Encoding

Foundational framework for probability distributions from binary representations. GPU-accelerated softmax integral computation for exact Monte Carlo inference.

Dec 2024

APOGÉE: Scaling Laws for Crypto Market Forecasting

Quantifying predictability limits in crypto markets. Scaling experiments across model size, dataset scope, and compute. Open benchmark for time-series forecasting.

Feb 2025
METHODOLOGY

RESEARCH PRINCIPLES

OPEN-SOURCE ETHOS

  • Core libraries and tools released under open licenses
  • Reproducible pipelines and training scripts
  • Modular design for community reuse and benchmarking

CALIBRATED EVALUATION

  • Probabilistic scoring metrics (extropy, CRPS, proper scoring rules)
  • Multi-horizon evaluation for short- and long-term reliability
  • Time-series aware validation with no look-ahead bias

SCIENTIFIC ENGINEERING

  • Built for integration: APIs, SDKs, and CLI tools
  • GPU-accelerated by default for scalability
  • Infrastructure tested under real-world constraints
05 COLLABORATE

Research Collaboration

We partner with researchers building world models for markets.