CUET-QwQ-32B — Question Generation Task

Is this run manual or automatic?: automatic
Is this run based on the provided starter kit?: no
Briefly describe this run: This run loads the TREC 2025 topics dataset and applies a 4-bit quantized version of the UnsLoTh QwQ-32B language model to generate ten critical investigative questions per news article. The questions aim to evaluate the trustworthiness of the articles by focusing on source bias, motivation, diversity of viewpoints, and factual accuracy. A carefully crafted prompt with few-shot examples guides the model. The output is parsed to extract unique questions, with multiple attempts per topic to ensure completeness. Finally, the results are formatted into a submission file for further use.
What other datasets or services (e.g. Google/Bing web search, ChatGPT, Perplexity, etc.) were used in producing the run?: No external search engines, APIs, or other language models such as ChatGPT or Perplexity were utilized in this run.
Briefly describe LLMs used for this run (optional): The run uses the unsloth/QwQ-32B-Preview-bnb-4bit model, a large-scale causal language model optimized via 4-bit quantization to reduce memory requirements while retaining strong generation capabilities. This model is loaded through the UnsLoTh framework and integrated with HuggingFace pipelines and LangChain for prompt management and generation.
Please give this run a priority for inclusion in manual assessments.: 5 (bottom)