Open Issues Need Help
View All on GitHub Design an intervention test for world models 15 days ago
help wanted research
physicalreasoning/forward physicalreasoning/forward Sealed forward tests for world models: entries are frozen before a round and scored on outcomes that resolve after it.
Open project →
0
Sealed forward tests for world models: entries are frozen before a round and scored on outcomes that resolve after it.
#benchmark#evaluation#forecasting#prediction-markets#proper-scoring-rules#world-models
help wanted research
physicalreasoning/forward physicalreasoning/forward Sealed forward tests for world models: entries are frozen before a round and scored on outcomes that resolve after it.
Open project →
0
Sealed forward tests for world models: entries are frozen before a round and scored on outcomes that resolve after it.
#benchmark#evaluation#forecasting#prediction-markets#proper-scoring-rules#world-models
good first issue infra
physicalreasoning/forward physicalreasoning/forward Sealed forward tests for world models: entries are frozen before a round and scored on outcomes that resolve after it.
Open project →
0
Sealed forward tests for world models: entries are frozen before a round and scored on outcomes that resolve after it.
#benchmark#evaluation#forecasting#prediction-markets#proper-scoring-rules#world-models