Benchmark Design

2 papers touch this idea.

Papers

AIs that play card games fail because they forget what they saw, not because they choose badly

Answer first, check later: AI cuts video-search cost by 93%

Related concepts

Ideas that show up alongside Benchmark Design.

Activities Of Daily Living Understanding Closed-Loop Policy Evaluation Efficient Inference Fine-Tuning On Policy Rollouts Large Language Model Reasoning Long Video Question Answering Memory And Observation Recall Multimodal Large Language Models Non-Markov Games Spatial Reasoning Temporal Grounding Temporal Reasoning Vision Language Models