Do Androids Know They're Only Dreaming of Electric Sheep?
December 28, 2023 ยท Declared Dead ยท ๐ Annual Meeting of the Association for Computational Linguistics
"No code URL or promise found in abstract"
Evidence collected by the PWNC Scanner
Authors
Sky CH-Wang, Benjamin Van Durme, Jason Eisner, Chris Kedzie
arXiv ID
2312.17249
Category
cs.CL: Computation & Language
Cross-listed
cs.AI,
cs.LG
Citations
60
Venue
Annual Meeting of the Association for Computational Linguistics
Last Checked
2 months ago
Abstract
We design probes trained on the internal representations of a transformer language model to predict its hallucinatory behavior on three grounded generation tasks. To train the probes, we annotate for span-level hallucination on both sampled (organic) and manually edited (synthetic) reference outputs. Our probes are narrowly trained and we find that they are sensitive to their training domain: they generalize poorly from one task to another or from synthetic to organic hallucinations. However, on in-domain data, they can reliably detect hallucinations at many transformer layers, achieving 95% of their peak performance as early as layer 4. Here, probing proves accurate for evaluating hallucination, outperforming several contemporary baselines and even surpassing an expert human annotator in response-level detection F1. Similarly, on span-level labeling, probes are on par or better than the expert annotator on two out of three generation tasks. Overall, we find that probing is a feasible and efficient alternative to language model hallucination evaluation when model states are available.
Community Contributions
Found the code? Know the venue? Think something is wrong? Let us know!
๐ Similar Papers
In the same crypt โ Computation & Language
๐
๐
Old Age
๐
๐
Old Age
BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding
R.I.P.
๐ป
Ghosted
Language Models are Few-Shot Learners
R.I.P.
๐ป
Ghosted
RoBERTa: A Robustly Optimized BERT Pretraining Approach
R.I.P.
๐ป
Ghosted
BART: Denoising Sequence-to-Sequence Pre-training for Natural Language Generation, Translation, and Comprehension
R.I.P.
๐ป
Ghosted
Deep contextualized word representations
Died the same way โ ๐ป Ghosted
R.I.P.
๐ป
Ghosted
PyTorch: An Imperative Style, High-Performance Deep Learning Library
R.I.P.
๐ป
Ghosted
XGBoost: A Scalable Tree Boosting System
R.I.P.
๐ป
Ghosted
Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift
R.I.P.
๐ป
Ghosted