BabyTheorist: A Benchmark for Learning to Theorize the World from Observation Alone
cs.AI
Submitted: 2026-11-05
Updated: 2026-11-06
Code: https://github.com/ahn-ml/babytheorist
Terminology
Sources
- On the Measure of Intelligence
- ARC Prize 2024: Technical Report
- ARC Prize 2025: Technical Report
- Bongard-LOGO: A New Benchmark for Human-Level Concept Learning and Reasoning
- Bongard-HOI: Benchmarking Few-Shot Visual Reasoning for Human-Object Interactions
- Learning to Theorize the World from Observation
- COGITAO: A Visual Reasoning Framework To Study Compositionality & Generalization
- COGS: A Compositional Generalization Challenge Based on Semantic Interpretation
- CLEVRER: CoLlision Events for Video REpresentation and Reasoning
- DreamCoder: Growing generalizable, interpretable knowledge with wake-sleep Bayesian program learning
- Learning Latent Dynamics for Planning from Pixels
- Dream to Control: Learning Behaviors by Latent Imagination
- Mastering Atari with Discrete World Models
- Mastering Diverse Domains through World Models
- Latent Action Pretraining from Videos
- Genie: Generative Interactive Environments
- Searching Latent Program Spaces
- Factored Latent Action World Models
- LoRA: Low-Rank Adaptation of Large Language Models
- Proximal Policy Optimization Algorithms
Related papers
- Multi-Objective Structured Pruning of LLMs for Latency and Model Size Optimization
- MAVEN-T: Reinforced Heterogeneous Distillation for Real-Time Multi-Agent Trajectory Prediction
- Model Discovery Agent: LLM-assisted Bayesian experiment design for data-efficient discovery of mechanistic world models
- Embeddings for Preferences, Not Semantics
- Can We Trust Item Response Theory for AI Evaluation?
- Diagnosing Task Insensitivity in Language Agents