Rainbow-Padding We introduce Rainbow Padding, a cyclic multi-token padding scheme that eliminates early termination and restores length robustness in instruction-tune quasar529/rainbow-padding-llada Text Generation • Updated Oct 9 • 32
Model with SAFEPATH AI-ISL/DeepSeek-R1-Distill-Qwen-7B-SP Text Generation • 8B • Updated May 27 • 9 AI-ISL/DeepSeek-R1-Distill-Llama-8B-SP Text Generation • 8B • Updated May 27 • 13 AI-ISL/HarmChain Viewer • Updated May 27 • 3.72k • 17 • 2
R-TOFU: Unlearning in Large Reasoning Models sangyon/R-TOFU Viewer • Updated Jun 27 • 10.6k • 55 sangyon/Reasoned_IDK Viewer • Updated Jun 1 • 400 • 21 sangyon/LRM-target 8B • Updated Apr 20 • 9
DUSK: Do not Unlearn Shared Knowledge AI-ISL/DUSK-target 8B • Updated Apr 26 • 75 • 3 AI-ISL/DUSK-retrain 8B • Updated May 3 • 7 AI-ISL/DUSK Viewer • Updated May 16 • 856 • 428 • 1
Rainbow-Padding We introduce Rainbow Padding, a cyclic multi-token padding scheme that eliminates early termination and restores length robustness in instruction-tune quasar529/rainbow-padding-llada Text Generation • Updated Oct 9 • 32
R-TOFU: Unlearning in Large Reasoning Models sangyon/R-TOFU Viewer • Updated Jun 27 • 10.6k • 55 sangyon/Reasoned_IDK Viewer • Updated Jun 1 • 400 • 21 sangyon/LRM-target 8B • Updated Apr 20 • 9
Model with SAFEPATH AI-ISL/DeepSeek-R1-Distill-Qwen-7B-SP Text Generation • 8B • Updated May 27 • 9 AI-ISL/DeepSeek-R1-Distill-Llama-8B-SP Text Generation • 8B • Updated May 27 • 13 AI-ISL/HarmChain Viewer • Updated May 27 • 3.72k • 17 • 2
DUSK: Do not Unlearn Shared Knowledge AI-ISL/DUSK-target 8B • Updated Apr 26 • 75 • 3 AI-ISL/DUSK-retrain 8B • Updated May 3 • 7 AI-ISL/DUSK Viewer • Updated May 16 • 856 • 428 • 1
