mem0ai / mem0ai/memory-benchmarks
Question: Is the LoCoMo `_JUDGE_TEMPLATE` too lenient for answer grading?
Nobody has claimed this yet.
- Dominant language
- Python
- Stars
- 112
- Forks
- 46
- PR merge metrics
- No merged PRs in 30d
Description
Hi, thanks for open-sourcing the benchmark implementation.
I had a question about the LoCoMo judge prompt in benchmarks/locomo/prompts.py, especially _JUDGE_TEMPLATE.
Some of the current grading rules seem intentionally permissive, for example:
- Partial credit marks an answer as
CORRECTif it includes at least one correct item from the gold answer list. - Evidence is used only to accept answers, not to reject them more strictly.
- Same named entity / same referent can be marked
CORRECTeven if the generated answer provides different details. - The prompt says to mark
WRONGonly when the generated answer contains zero correct items from the gold answer or addresses a completely different topic. - Dates within 14 days and durations within 50% are accepted.
In our local runs, this appears to make the judge accept some answers that are only partially correct or that miss key details, which may inflate accuracy by turning false positives into CORRECT labels.
Could you clarify whether this level of leniency is intended for the reported LoCoMo scores? Or would you consider adding a stricter judge mode / prompt variant for cases where the generated answer should match all required answer items and avoid conflicting extra details?
A stricter alternative might require:
- list/count questions to include all required items, not just one;
- extra details to be accepted only if they do not conflict with the gold answer or evidence;
- evidence to be usable for both accepting and rejecting answers;
- same referent to be insufficient when the question asks for a specific attribute, date, action, or reason.
Happy to provide concrete false-positive examples if useful.
Contributor guide
No contributing guide indexed for this repository
First steps
- Read the whole issue, then the project's contributing guide.
- Comment on the issue to say you are picking it up — it saves two people doing the same work.
- Fork the repository and make your change on a branch.
- Open a pull request that references the issue number.
Research direction
Start with benchmarks/locomo/prompts.py and inspect _JUDGE_TEMPLATE, then review the current LoCoMo grading rules and any benchmark evaluation entry points. Compare the permissive cases described here with concrete false-positive examples. Done would require an agreed stricter grading behavior or mode, documented acceptance criteria, and tests showing how required items, conflicting details, evidence, and referents are handled.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- python
- Domain
- testing-qa
- Issue type
- Feature
- Difficulty
- 5/5
- Estimated time
- Over a week
- Activity status
- Quiet
- Clarity
- Mostly clear
- Newbie friendliness
- 35/100