Beware of Unreliable Data in Model Evaluation: A LLM Prompt Selection Case Study (cleanlab.ai)

Screenshot of cleanlab.ai: Beware of Unreliable Data in Model Evaluation: A LLM Prompt Selection Case Study

First screen, captured 3 years ago.

Looks similar

Screenshot of mlops.systems: How to think about creating a dataset for LLM fine-tuning evaluation

How to think about creating a dataset for LLM fine-tuning evaluation (mlops.systems)

120 points by sebg 2 years ago | 7 comments

Screenshot of scalarlm.com: LLM-Deflate: Extracting LLMs into Datasets

LLM-Deflate: Extracting LLMs into Datasets (scalarlm.com)

53 points by gdiamos 1 year ago | 26 comments

Screenshot of nuenki.app: Experimentation Matters: Why Nuenki isn't using pairwise evaluations

Experimentation Matters: Why Nuenki isn't using pairwise evaluations (nuenki.app)

4 points by Alex-Programs 1 year ago | 0 comments

Screenshot of trygloo.com: Classifying customer messages with LLMs vs traditional ML

Classifying customer messages with LLMs vs traditional ML (trygloo.com)

237 points by hellovai 3 years ago | 102 comments

Screenshot of lcamtuf.substack.com: LLMs are good at playing you

LLMs are good at playing you (lcamtuf.substack.com)

117 points by zdw 3 years ago | 56 comments

Screenshot of efraingaray.com: TabPFN and TabICL vs. tuned XGBoost: the model that doesn't train won 14/14

TabPFN and TabICL vs. tuned XGBoost: the model that doesn't train won 14/14 (efraingaray.com)

21 points by EfrainGaray 28 days ago | 9 comments

Blog post benchmarking TabPFN and TabICL foundation models against tuned XGBoost on fourteen tabular datasets, finding the no-training models won all fourteen.

tabpfn tabicl xgboost tabular-data

Looks similar

How to think about creating a dataset for LLM fine-tuning evaluation(mlops.systems)

LLM-Deflate: Extracting LLMs into Datasets(scalarlm.com)

Experimentation Matters: Why Nuenki isn't using pairwise evaluations(nuenki.app)

Classifying customer messages with LLMs vs traditional ML(trygloo.com)

LLMs are good at playing you(lcamtuf.substack.com)

TabPFN and TabICL vs. tuned XGBoost: the model that doesn't train won 14/14(efraingaray.com)