MM-Eureka: Exploring Visual Aha Moment with Rule-based Large-scale Reinforcement Learning Paper • 2503.07365 • Published 4 days ago • 53
ainbo/text_and_concat_image_hf_version_epoch_1_with_prefix_with_exist_split_fixed_best_of_16_CoT Viewer • Updated 4 days ago • 22.9k • 62
ainbo/_only_thought_text_hf_version_epoch_1_with_prefix_with_exist_split_fixed_best_of_16_images Viewer • Updated 4 days ago • 9.18k • 161
ainbo/xt_hf_version_epoch_1_with_prefix_with_exist_split_fixed_best_of_16_mix_thought_and_images Viewer • Updated 4 days ago • 9.18k • 162
ainbo/dreambench_eval_results_seed_x_normal_dreambench_best_of_16_images Viewer • Updated 4 days ago • 9.18k • 163
ainbo/t_and_concat_image_hf_version_epoch_1_with_prefix_with_exist_split_fixed_best_of_16_images Viewer • Updated 4 days ago • 9.19k • 159
ainbo/h_exist_split_fixed_best_of_16_mix_thought_and_images_lm_loss_scale_3_0_rec_loss_scale_6_0 Viewer • Updated 4 days ago • 7.49k • 116
ainbo/t_and_concat_image_hf_version_epoch_1_with_prefix_with_exist_split_fixed_best_of_16_images Viewer • Updated 4 days ago • 9.19k • 159
ainbo/cobsat_eval_results_seedx_cot_with_finetune_self_collect_with_thought_pass_16 Viewer • Updated 4 days ago • 40k • 47
ainbo/h_exist_split_fixed_best_of_16_mix_thought_and_images_lm_loss_scale_3_0_rec_loss_scale_6_0 Viewer • Updated 4 days ago • 7.49k • 116
ainbo/cobsat_eval_results_seedx_default_best_of_N_pass_16_of_images Viewer • Updated 4 days ago • 40k • 63
ainbo/_only_thought_text_hf_version_epoch_1_with_prefix_with_exist_split_fixed_best_of_16_images Viewer • Updated 4 days ago • 9.18k • 161
ainbo/xt_hf_version_epoch_1_with_prefix_with_exist_split_fixed_best_of_16_mix_thought_and_images Viewer • Updated 4 days ago • 9.18k • 162
ainbo/text_and_concat_image_hf_version_epoch_1_with_prefix_with_exist_split_fixed_best_of_16_CoT Viewer • Updated 4 days ago • 22.9k • 62
ainbo/cobsat_eval_results_seedx_cot_finetune_with_thought_and_concat_image_best_of_N_pass_16_of_ Viewer • Updated 4 days ago • 7.78k • 61
ainbo/cobsat_eval_results_seedx_cot_finetune_with_only_thought_best_of_N_pass_16_of_thought_and_ Viewer • Updated 4 days ago • 40k • 64
ainbo/cobsat_eval_results_seedx_cot_finetune_with_only_thought_best_of_N_pass_16_of_images Viewer • Updated 4 days ago • 40k • 75
ainbo/dreambench_eval_results_seed_x_normal_dreambench_best_of_16_images Viewer • Updated 4 days ago • 9.18k • 163
ainbo/dreambench_eval_results_seed_cot_of_InternVL2_5_78b_mpo_awq_cot_with_two_component Viewer • Updated 4 days ago • 2.06k • 43