Collection of models and datasets for Beyond Binary Rewards: Training LMs to Reason about their Uncertainty
Mehul Damani PRO
mehuldamani
AI & ML interests
Reinforcement Learning, Large Language Models
Recent Activity
published
a dataset about 23 hours ago
mehuldamani/multi-answer-sft-target-dataset published
a model 2 days ago
mehuldamani/sfted_rlvr_multi__veryHardDataset_moreThinking updated
a dataset 2 days ago
mehuldamani/multi-answer-sft-target-dataset Organizations
None yet
models 218
mehuldamani/sfted_rlvr_multi__veryHardDataset_moreThinking
Updated
mehuldamani/qwen25Instruct_rlvr_single_veryHardDataset
Updated
mehuldamani/qwen25Instruct_rlvr_multi_veryHardDataset_moreThinking
Updated
mehuldamani/qwen25Instruct_rlvr_multi_hardDataset_moreThinking
Updated
mehuldamani/sft-base-half-tranches-v1-global-step-394
Text Generation • 8B • Updated
• 1.73k
mehuldamani/qwen25Instruct_rlvr_single_hardDataset
Updated
mehuldamani/qwen25Base_rlvr_multi_hardDataset
Updated
mehuldamani/sftedModelBase_rlvr_multi_hardDataset
Updated
mehuldamani/qwen25Instruct_rlvr_multi_hardDataset
Updated
mehuldamani/regularBrier_mixedNumCandidates_rlcr_multi_from_rlvr_chkpt360
Updated
datasets 56
mehuldamani/multi-answer-sft-target-dataset
Viewer
• Updated
• 1.59k • 6
mehuldamani/big-math-very-tough
Viewer
• Updated
• 12.5k • 31 • 1
mehuldamani/hotpot_qa_test_gold_removed_1
Viewer
• Updated
• 20.5k • 9
mehuldamani/hotpot_qa_test_gold_removed_2
Viewer
• Updated
• 20.5k • 9
mehuldamani/hotpot_qa_trainTest_gold_removed_2
Viewer
• Updated
• 20.5k • 10
mehuldamani/hotpot2Removed_eval_10Runs_rlvr_multi_on_rlcr_multi
Viewer
• Updated
• 500 • 7
mehuldamani/big-math-tough
Viewer
• Updated
• 18.5k • 44
mehuldamani/medTroubleshootig-rlvr-220-evaled-on-rlcr
Viewer
• Updated
• 5k • 6
mehuldamani/medTroubleshootig-rlvr-220-evaled-on-rlvr
Viewer
• Updated
• 5k • 4
mehuldamani/medDataset_25k
Viewer
• Updated
• 75k • 11