task complexity
An assessment of how challenging a specific task is for an AI model, often based on the number of components, requirements for reasoning, and the diversity of scenarios to be addressed.
- Constructing an Optimal Behavior Basis for the Option Keyboard
- Each Complexity Deserves a Pruning Policy
- Eluder dimension: localise it!
- LabUtopia: High-Fidelity Simulation and Hierarchical Benchmark for Scientific Embodied Agents
- LooGLE v2: Are LLMs Ready for Real World Long Dependency Challenges?
- Measuring and Controlling Solution Degeneracy across Task-Trained Recurrent Neural Networks
- NeuroPath: Neurobiology-Inspired Path Tracking and Reflection for Semantically Coherent Retrieval
- PhysGym: Benchmarking LLMs in Interactive Physics Discovery with Controlled Priors
- Quantifying Generalisation in Imitation Learning
- TheAgentCompany: Benchmarking LLM Agents on Consequential Real World Tasks
- To Think or Not To Think: A Study of Thinking in Rule-Based Visual Reinforcement Fine-Tuning