DRIFT: Learning from Abundant User Dissatisfaction in Real-World Preference Learning Paper • 2510.02341 • Published Sep 27, 2025 • 3
AmberYifan/Qwen2.5-7B-Instruct-wildfeedback-DRIFT-iter2 Text Generation • 8B • Updated Jun 5, 2025 • 6
AmberYifan/Qwen2.5-7B-Instruct-wildfeedback-iterDPO-iter2 Text Generation • 8B • Updated Jun 25, 2025 • 7 • 1
AmberYifan/Qwen2.5-7B-Instruct-wildfeedback-SPIN-iter2 Text Generation • 8B • Updated Jun 21, 2025 • 12 • 1
AmberYifan/Qwen2.5-14B-Instruct-ultrafeedback-iterdpo-iter2-RPO Text Generation • 841k • Updated Aug 7, 2025 • 1
AmberYifan/Qwen2.5-14B-Instruct-ultrafeedback-DRIFT-iter2-RPO Text Generation • 841k • Updated Aug 7, 2025
AmberYifan/Qwen2.5-14B-Instruct-ultrafeedback-spin-iter2-RPO Text Generation • 841k • Updated Aug 7, 2025 • 3
AmberYifan/Qwen2.5-14B-Instruct-wildfeedback-RPO-DRIFT-iter2 Text Generation • 841k • Updated Jul 30, 2025
AmberYifan/Qwen2.5-14B-Instruct-wildfeedback-RPO-iterDPO-iter2 Text Generation • 841k • Updated Jul 30, 2025
AmberYifan/Qwen2.5-14B-Instruct-wildfeedback-RPO-SPIN-iter2 Text Generation • 841k • Updated Jul 30, 2025
AmberYifan/Qwen2.5-7B-Instruct-ultrafeedback-iterDPO-iter2 Text Generation • 8B • Updated Jun 30, 2025 • 4 • 1
AmberYifan/Qwen2.5-7B-Instruct-ultrafeedback-DRIFT-iter2 Text Generation • 8B • Updated Jun 29, 2025 • 1
AmberYifan/Qwen2.5-7B-Instruct-ultrafeedback-SPIN-iter2 Text Generation • 8B • Updated Jun 30, 2025 • 1 • 1