Skip to content

Pull requests: IBM/unitxt

New pull request New
33 Open 1,667 Closed
33 Open 1,667 Closed
Author
Filter by author
Loading
Label
Filter by label
Loading
Use alt + click/return to exclude labels
or ⇧ + click/return for logical OR
Projects
Filter by project
Loading
Milestones
Filter by milestone
Loading
Reviews
Filter by reviews
No reviews Review required Approved review Changes requested
Assignee
Filter by who’s assigned
Assigned to nobody Loading
Sort
Sort by
Newest Oldest Most commented Least commented Recently updated Least recently updated Best match
Most reactions
👍 👎 😄 🎉 😕 ❤️ 🚀 👀

Pull requests list

New metric definitions for llama-3-3-70b as judge in Arena Hard benchmark
#1949 opened Oct 27, 2025 by kmazrolina Loading…
Rag metric update again
#1948 opened Oct 21, 2025 by dafnapension Loading…
1
Add format and system prompt to task meta data
#1947 opened Oct 15, 2025 by yoavkatz Loading…
Update rag llm as judge metric to support llama-3-3-70b model on WML.
#1945 opened Oct 9, 2025 by piotrhelm Loading…
2
Safety benchmark
#1943 opened Oct 3, 2025 by bnayahu Loading…
Enable summarization by subsets and groups
#1942 opened Oct 3, 2025 by bnayahu Loading…
Correct reflection based tool calling metrics so valid results will be 1.
#1940 opened Sep 25, 2025 by yoavkatz Loading…
light fast removal of register_all_artifacts for unitxt classes
#1939 opened Sep 20, 2025 by dafnapension Loading…
4
Configure Mend for GitHub Enterprise
#1937 opened Sep 18, 2025 by ibm-mend-app bot Loading…
Add EvalAssist judges integration
#1936 opened Sep 16, 2025 by martinscooper • Draft
A potential fix for messy error box when Error is KeyError
#1932 opened Sep 1, 2025 by dafnapension Loading…
6
Refactor, simplify and unify llm as judges to map reduce metric
#1927 opened Aug 24, 2025 by elronbandel • Draft
fixed mmmu by cooking options from answer, when options is not given in the instance
#1921 opened Aug 20, 2025 by dafnapension Loading…
Fix bfcl so that instances do not explode in size, and all cards yield recipes that pass _source_to_dataset
#1915 opened Aug 13, 2025 by dafnapension Loading…
1
Refactor inference
#1914 opened Aug 13, 2025 by elronbandel Loading…
Add vllm to cross provider engine
#1910 opened Aug 7, 2025 by elronbandel Loading…
Allow using python functions instead of operators (e.g in pre-processing pipeline)
#1845 opened Jun 26, 2025 by elronbandel Loading…
1
Ccc inference
#1745 opened Apr 21, 2025 by eladven • Draft
For issue 1575: Eliminating Manual Class Registration in Unitxt, replaced by Import Paths
#1713 opened Apr 5, 2025 by dafnapension Loading…
4
Add audio support
#1706 opened Apr 1, 2025 by elronbandel Loading…
Unify llm judges into a single prepare file
#1696 opened Mar 21, 2025 by martinscooper Loading…
5
Momentary fix of 'CriteriaWithOptions is not JSON serializable'
#1694 opened Mar 19, 2025 by martinscooper • Draft
BugFix: Use dumping of task data and source only when dumping
#1661 opened Mar 9, 2025 by elronbandel Loading…
1
Add batch size control to Huggingface pipeline based inference engine
#1636 opened Feb 27, 2025 by elronbandel Loading…
Add prediction variable name customization to LLM as Judge
#1622 opened Feb 21, 2025 by martinscooper • Draft
4
ProTip! Updated in the last three days: updated:>2025-10-24.