Datasets:

allenai
/

olmOCR-bench

✖ Invalid input: expected object, received undefined → at tasks[0].field_spec ✖ Invalid input: expected array, received undefined → at tasks[0].solvers ✖ Invalid input: expected array, received undefined → at tasks[0].scorers ✖ Invalid input: expected object, received undefined → at tasks[1].field_spec ✖ Invalid input: expected array, received undefined → at tasks[1].solvers ✖ Invalid input: expected array, received undefined → at tasks[1].scorers ✖ Invalid input: expected object, received undefined → at tasks[2].field_spec ✖ Invalid input: expected array, received undefined → at tasks[2].solvers ✖ Invalid input: expected array, received undefined → at tasks[2].scorers ✖ Invalid input: expected object, received undefined → at tasks[3].field_spec ✖ Invalid input: expected array, received undefined → at tasks[3].solvers ✖ Invalid input: expected array, received undefined → at tasks[3].scorers ✖ Invalid input: expected object, received undefined → at tasks[4].field_spec ✖ Invalid input: expected array, received undefined → at tasks[4].solvers ✖ Invalid input: expected array, received undefined → at tasks[4].scorers ✖ Invalid input: expected object, received undefined → at tasks[5].field_spec ✖ Invalid input: expected array, received undefined → at tasks[5].solvers ✖ Invalid input: expected array, received undefined → at tasks[5].scorers ✖ Invalid input: expected object, received undefined → at tasks[6].field_spec ✖ Invalid input: expected array, received undefined → at tasks[6].solvers ✖ Invalid input: expected array, received undefined → at tasks[6].scorers

Files changed (1) hide show

eval.yaml +1 -1

eval.yaml CHANGED Viewed

@@ -1,6 +1,6 @@
 name: olmOCR-bench
 description: olmOCR-bench is a benchmark to evaluate how well VLLMs can parse PDF and scanned documents into clean, semantic markdown/HTML.
-evaluation_framework: olmocr-bench
 tasks:
   - id: arxiv_math

 name: olmOCR-bench
 description: olmOCR-bench is a benchmark to evaluate how well VLLMs can parse PDF and scanned documents into clean, semantic markdown/HTML.
+evaluation_framework: inspect-ai
 tasks:
   - id: arxiv_math