# RAG-QA Challenge

This competition is a RAG task that can only be solved by using the fictional documents provided in the corpus.
AI tools may be used freely, but solution models are limited to the 27B size.
Participants should answer the questions in `test.jsonl` and submit predictions following the format in `sample_submission.jsonl`.

# Description

1. Validation Dataset (50 examples)
- File name: `dev.jsonl`
- `question`: The question to answer
- `id`: The question ID
- `reasoning_type`: The reasoning category (`contextual_single_hop`, `causal_chain`, `bridge_multi_doc`, `unanswerable`, `temporal_resolution`, `comparison_with_filtering`, `conflict_resolution`, `aggregation`, `bridge_multi_sentence`)
- `evidence`: The document and sentence references needed to answer the question, e.g. `[{"doc_id": "", "sent_id": 1}]`
- `answers`: The gold answer label(s)

2. Test Dataset (1,000 examples)
- File name: `test.jsonl`

3. Corpus Dataset (2,253 documents)
- File name: `corpus.jsonl`

4. Submission Format
- File name: `sample_submission.jsonl`

# Files

```text
corpus.jsonl
dev.jsonl
test.jsonl
sample_submission.jsonl
score_submission.py
```