Teaching language models to think efficiently with Adaptive Length Penalty (ALP)
AI & ML interests
Scaling up good synthetic reasoning. Post-training and synthetic data research lab.
Organization Card
SynthLabs
Advancing and Scaling Synthetic Reasoning through Post-Training AI Research
This collection contains assets associated with the Big-Math dataset, a high-quality collection of over 250,000 math questions with verifiable answers
-
SynthLabsAI/Big-Math-RL-Verified
Viewer β’ Updated β’ 251k β’ 8.42k β’ 222 -
SynthLabsAI/Big-Math-RL-UNVERIFIED
Viewer β’ Updated β’ 34.9k β’ 9 β’ 1 -
nlile/NuminaMath-1.5-RL-Verifiable
Viewer β’ Updated β’ 131k β’ 5.99k β’ 9 -
Big-Math: A Large-Scale, High-Quality Math Dataset for Reinforcement Learning in Language Models
Paper β’ 2502.17387 β’ Published β’ 7
Teaching language models to think efficiently with Adaptive Length Penalty (ALP)
This collection contains assets associated with the Big-Math dataset, a high-quality collection of over 250,000 math questions with verifiable answers
-
SynthLabsAI/Big-Math-RL-Verified
Viewer β’ Updated β’ 251k β’ 8.42k β’ 222 -
SynthLabsAI/Big-Math-RL-UNVERIFIED
Viewer β’ Updated β’ 34.9k β’ 9 β’ 1 -
nlile/NuminaMath-1.5-RL-Verifiable
Viewer β’ Updated β’ 131k β’ 5.99k β’ 9 -
Big-Math: A Large-Scale, High-Quality Math Dataset for Reinforcement Learning in Language Models
Paper β’ 2502.17387 β’ Published β’ 7
datasets 6
SynthLabsAI/Big-Math-RL-Verified
Viewer
β’ Updated
β’ 251k β’ 8.42k β’ 222
SynthLabsAI/Big-Math-RL-UNVERIFIED
Viewer
β’ Updated
β’ 34.9k β’ 9 β’ 1
SynthLabsAI/PERSONA
Viewer
β’ Updated
β’ 200k β’ 3.44k β’ 18
SynthLabsAI/PERSONA_subset
Viewer
β’ Updated
β’ 5k β’ 3.42k β’ 3
SynthLabsAI/PRISM-Filter
Viewer
β’ Updated
β’ 3.87k β’ 8
SynthLabsAI/Synthetic-Personas
Viewer
β’ Updated
β’ 1k β’ 9 β’ 3