Hello Zindians,Is there any person who successfully implemented RL finetuning on this dataset and is willing to share his/her approach (Code would be appreciate... See more

.

Hello everyone,We have now finalised the evaluation process for the challenge. Thank you all for your hard work and participation throughout this journey.In lin... See more

I just wonder how the moderator or judge will ensure a fair result across all teams as they mentioned that they will only use the trained model and after the co... See more

Dear participants,Thank you all for the great effort. We are very pleased about your engament and looking forward to read about your work. In order to get to th... See more

It's almost impossible to train models and got good score from scratch in such a short period. I feel it's unfaire for people who started building solutions sin... See more

fine tunning for general questions will not work on unseen data

Hello everyone 👋 Just a quick reminder that submissions with answers copied across tracks will be disqualified.Please also note that top performers in each ind... See more

Hi @AntonioDeDomenico, Please, could you clarify the rules for data generation, especially for the new problem types, and whether we can use this data for model... See more

Hi,When I try to submit my job to the competition platform to fine-tune the LLM, I consistently get this error:502 Bad Gateway returned by CloudflareI’ve tried... See more

What tricks did you use to manage the three models given their large parameters? I'm trying to work on my first submission, but I'm unable to proceed because my... See more

Hi @AntonioDeDomenico, It seems that some participants are fine-tuning on the test set. How do the organizers plan to detect models trained on the test data? If... See more

The Public Leaderboard is scheduled to close on February 2nd, with the Private Leaderboard evaluation to follow. Currently, only model predictions are being sub... See more

I hope you’re doing well.I wanted to ask if it would be possible to get a short extension on the submission deadline. Over the past weeks, the internet in Iran... See more

The mentioned closing time is 02 Feb 2026 at 04:59. But the mentioned cloding time is in which timezone.

I had a quick question to better understand how fairness is being ensured for all participants. For top submissions, do you re-run the training using the submit... See more

Hi, Thankyou for the competition. I wanted to clarify two things related to datasets: 1. Are we allowed to use publicly available datasets for the competition,... See more

Is it possible to perform fine-tuning with the phase 02 test data?

When submitting the final submission, is it possible to only submit the final labeled answers, or is it necessary to submit the complete LLM output, considering... See more

Hi, It is sighlty unclear from the previous discussions that is it allowed to use test samples to generate more data for training or not, it will be heplful if... See more

Thank you for organizing the competition.I found that one team member hasn't join the team now, but the server said I can't invite the member because the compet... See more

I am getting this error : Qwen25 1.5b instruct error: Missing entries for IDs ID_AWAWP1ENMO_2, ID_2FT3OI6DBP_3, ID_7YEWB4HTUH_3, ID_D67V9OW6WB_3, ID_CJXNHM19NC_... See more

The submission gets a score of 0 for the Phase 1 portion of the dataset, despite receiving a valid score for the Phase 2 portion. We have verified that the data... See more

I have a quick question about the competition rules: Are prizes awarded only to the 1st place in each track (Track 1 / Track 2 / Track 3), meaning there are 3 t... See more

getting error - Qwen3 32b unknown error: empty output