TREC DL and LLM AggreFact experiments for relevance benchmark + prompts comparisons and groundedness vs Bespoke Minicheck 7B #1660
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
|
Check out this pull request on See visual diffs & provide feedback on Jupyter Notebooks. Powered by ReviewNB |
sfc-gh-dhuang
force-pushed
the
trec-exp-adding-rubric
branch
from
222f201 to
55f6564
Compare
sfc-gh-dhuang
force-pushed
the
trec-exp-adding-rubric
branch
from
55f6564 to
17941b0
Compare
sfc-gh-dhuang
changed the title
dosubot
bot
added
the
size:XXL
sfc-gh-dhuang
changed the title
sfc-gh-dhuang
changed the title
| @@ -0,0 +1,1017 @@ | |||
| { | |||
There was a problem hiding this comment.
Line #7. def scoreddocs_qrels_confusion_matrix(
Can you break this function up? A lot going on here, hard to follow
Reply via ReviewNB
...rk/benchmark_frameworks/experiments/context_relevance/trec_dl_passage_relevance_judges.ipynb
Show resolved
Hide resolved
...rk/benchmark_frameworks/experiments/context_relevance/trec_dl_passage_relevance_judges.ipynb
Show resolved
Hide resolved
| @@ -0,0 +1,1017 @@ | |||
| { | |||
There was a problem hiding this comment.
Line #8. class CustomTermFeedback(Provider):
Should we have OOB feedbacks for these metrics?
Reply via ReviewNB
...rk/benchmark_frameworks/experiments/context_relevance/trec_dl_passage_relevance_judges.ipynb
Show resolved
Hide resolved
sfc-gh-dhuang
force-pushed
the
trec-exp-adding-rubric
branch
from
8cabc46 to
9514bce
Compare
sfc-gh-jreini
approved these changes
Dec 10, 2024
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Description
Contain 2 main notebooks
trec_dl_passage_relevance_judges.ipynb: Experiments code with TREC DL passage level agreement and comparisons with Cortex team's hand-crafted prompt, UMBRELA prompt, and categorical prompt for search relevance LLM judges.Doc with details: https://docs.google.com/document/d/1lwNC11S3LnNHglDNzcJNgPsd7zpOGNwTNktn8pI3bUc/edit?tab=t.nnid3pbdfr33
llm_aggrefact_groundedness_benchmark.ipynb: Experiments code with LLM-AggreFact benchmark for fact-checking grounding documents and comparing Bespoke Minicheck-7B vs trulens groundedness (gpt-4o).Doc: https://docs.google.com/document/d/1oMWDlk6JK6HP3IJQ_pjqB5KlHMtDHUKNoi5uTyWEYhk/edit?disco=AAABZ2Ar0vc
Other details good to know for developers
Please include any other details of this change useful for TruLens developers.
Type of change
not work as expected)
Important
Add functions for generating balanced datasets and TREC DL passage benchmarks, and enhance feedback handling to support additional data types.
generate_balanced_llm_aggrefact_benchmarkandgenerate_trec_dl_passage_benchmarkindataset_preprocessing.pyfor creating balanced datasets and TREC DL passage benchmarks.run()infeedback.pyto supportdictas a valid feedback function output type.run_score_generation_on_single_row()intru_benchmark_experiment.pytofloat('nan')for invalid results.defaultdictandir_datasetsindataset_preprocessing.py.This description was created by
for 35432e7. It will automatically update as commits are pushed.