Allenda/VIEScore2-results

VIEScore2 — raw evaluation results Raw result files behind every number reported in the VIEScore2 paper, for independent verification. Produced by run_eval.py and the analyze_*.py scripts in the accompanying code release, under the frozen evaluation protocol (configs/eval_protocol.json). Layout eval_results/ — main 1,300-example multi-source suite: VIEScore2, VIEScore2 (w/o GRPO), replication seeds, equal-compute controls, reward ablations (dev_b* = F_β sweeps)… See the full description on the dataset page: https://huggingface.co/datasets/Allenda/VIEScore2-results.

<iframe src="https://silytics.si/embed/item/9e02b431-c709-4c3d-b6f7-3f9e713e35c4" width="420" height="190" style="border:0" loading="lazy" title="Silytics entry"></iframe>

Silytics shows public metadata from Hugging Face and links to the original. Information may be incomplete or out of date. Last refreshed 18 October 2026. Licenses and terms of the original apply to the work itself. Not financial, legal or professional advice.