RAID: A Shared Benchmark for Robust Evaluation of Machine-Generated Text Detectors
Fuente:
arXiv
Salvato in:
| Autori principali: | , , , , , , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2024
|
| Soggetti: | |
| Accesso online: | |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866917689110298624 |
|---|---|
| author | Dugan, Liam Hwang, Alyssa Trhlik, Filip Ludan, Josh Magnus Zhu, Andrew Xu, Hainiu Ippolito, Daphne Callison-Burch, Chris |
| author_facet | Dugan, Liam Hwang, Alyssa Trhlik, Filip Ludan, Josh Magnus Zhu, Andrew Xu, Hainiu Ippolito, Daphne Callison-Burch, Chris |
| contents | Many commercial and open-source models claim to detect machine-generated text with extremely high accuracy (99% or more). However, very few of these detectors are evaluated on shared benchmark datasets and even when they are, the datasets used for evaluation are insufficiently challenging-lacking variations in sampling strategy, adversarial attacks, and open-source generative models. In this work we present RAID: the largest and most challenging benchmark dataset for machine-generated text detection. RAID includes over 6 million generations spanning 11 models, 8 domains, 11 adversarial attacks and 4 decoding strategies. Using RAID, we evaluate the out-of-domain and adversarial robustness of 8 open- and 4 closed-source detectors and find that current detectors are easily fooled by adversarial attacks, variations in sampling strategies, repetition penalties, and unseen generative models. We release our data along with a leaderboard to encourage future research. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2405_07940 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | RAID: A Shared Benchmark for Robust Evaluation of Machine-Generated Text Detectors Dugan, Liam Hwang, Alyssa Trhlik, Filip Ludan, Josh Magnus Zhu, Andrew Xu, Hainiu Ippolito, Daphne Callison-Burch, Chris Computation and Language I.2.7 Many commercial and open-source models claim to detect machine-generated text with extremely high accuracy (99% or more). However, very few of these detectors are evaluated on shared benchmark datasets and even when they are, the datasets used for evaluation are insufficiently challenging-lacking variations in sampling strategy, adversarial attacks, and open-source generative models. In this work we present RAID: the largest and most challenging benchmark dataset for machine-generated text detection. RAID includes over 6 million generations spanning 11 models, 8 domains, 11 adversarial attacks and 4 decoding strategies. Using RAID, we evaluate the out-of-domain and adversarial robustness of 8 open- and 4 closed-source detectors and find that current detectors are easily fooled by adversarial attacks, variations in sampling strategies, repetition penalties, and unseen generative models. We release our data along with a leaderboard to encourage future research. |
| title | RAID: A Shared Benchmark for Robust Evaluation of Machine-Generated Text Detectors |
| topic | Computation and Language I.2.7 |
| url | https://arxiv.org/abs/2405.07940 |