gsq
Datasets
All datasets matching “gsq”GS-QA2-source
GS-QA2 Source Data
Reference geospatial data underlying GS-QA2, a benchmark for question answering over raster–vector data. This repository contains the raw vector and raster inputs needed to rebuild the reference PostGIS database, execute the benchmark's ground-truth SQL queries, and run the baselines — use it together with the QA pairs in Zoe/GS-QA2 and the code in github.com/ZhuochengShang/QARV.
GS-QA2 resources: all resources collection · benchmark QA pairs · public… See the full description on the dataset page: https://huggingface.co/datasets/Zoe/GS-QA2-source.Qwen3.8-27B-GSQ-RCO-scale-recovery-evidence
Qwen3.8-27B GSQ-RCO IQ3_S: scale-recovery evidence and code
Research evidence dataset. No model weights. Part of the collection
Xyntetik Research: Model Surgery and Scale Recovery on this account, produced with
Xyntetik Runner.
Dataset summary
Question tested. Whether retraining only the fp16 block scales of an existing IQ3_S file (integer codes untouched) by distillation against the BF16 parent brings it inside the house bar, and what it does on a public… See the full description on the dataset page: https://huggingface.co/datasets/Joakimpalm-Zen/Qwen3.8-27B-GSQ-RCO-scale-recovery-evidence.spoken-alpaca-gpt4GS-QA2
GS-QA2: A Benchmark for Question Answering over Raster–Vector Data
Release scope: The original canonical release contains 3,300 public questions, and its four existing configurations and public test splits remain unchanged. The new_v1 extension publishes 300 development questions: 100 each in new_vector, new_raster_only, and new_raster_vector. A separate 150-question challenge (50 per track) is retained privately on the maintainers' lab cluster. Its questions, SQL, answers, and… See the full description on the dataset page: https://huggingface.co/datasets/Zoe/GS-QA2.NMSQA-test
Dataset Card for "NMSQA_test"
More Information needed
gsq