arxiv:2505.11711
sagnik mukherjee
sagnikM
AI & ML interests
None yet
Recent Activity
upvoted a paper about 1 month ago
Golden Goose: A Simple Trick to Synthesize Unlimited RLVR Tasks from Unverifiable Internet Text updated
a model about 1 month ago
sagnikM/grpo_rmsprop_qwen3-8b_3k_seqlen published
a model about 1 month ago
sagnikM/grpo_rmsprop_qwen3-8b_3k_seqlen Organizations
None yet
models 15
sagnikM/grpo_rmsprop_qwen3-8b_3k_seqlen
8B • Updated
sagnikM/grpo_rmsprop_llama3p1_8b_3k_seqlen_1e-7
Text Generation • 8B • Updated
• 5
sagnikM/grpo_rmsprop_qwen3_1p7b_3k_seqlen_1e-5
Text Generation • 2B • Updated
• 5
sagnikM/grpo_rmsprop_qwen3_1p7b_3k_seqlen_1e-6
Text Generation • 2B • Updated
• 5
sagnikM/grpo_sgd_qwen3-8b_3k_seqlen_momentum_0p9_1e-2
Text Generation • 8B • Updated
• 37
sagnikM/grpo_sgd_llama3p1_8b_3k-seqlen_momentum_0p9_1e-3
Text Generation • 8B • Updated
• 31
sagnikM/grpo_sgd_qwen3_1p7b_3k-seqlen_momentum_0p9_1e-2
Text Generation • 2B • Updated
• 2
sagnikM/grpo_sgd_qwen3_1p7b_3k-seqlen_momentum_0p9_1e-1
Text Generation • 2B • Updated
• 2
sagnikM/grpo_sgd_qwen3-8b_3k_seqlen
Text Generation • 8B • Updated
• 6
sagnikM/grpo_adam_qwen3-8b_3k_seqlen
Text Generation • 8B • Updated