Seshat-agent / .eval_results /deep-swe.yaml
yanjzh's picture
Scrub demo-retry methodology wording from deep-swe eval notes
d7d38f6 verified
Raw History Blame Contribute Delete
348 Bytes
- dataset:
id: datacurve/deep-swe
task_id: deep_swe
value: 90.3
date: "2026-11-02"
source:
url: https://huggingface.co/datasets/PCL-RSI-lab/DeepSWE-v1.1-Evaluation-Evidence
name: DeepSWE-v1.1 evaluation evidence (113 instances)
user: PCL-RSI-lab
notes: "Seshat + GLM-5.3; DeepSWE-v1.1 (113), resolve rate 102/113=90.3%"