3 |
Evaluation of Unsupervised Automatic Readability Assessors Using Rank Correlations ...
|
|
|
|
BASE
|
|
Show details
|
|
4 |
Analysis of Language Change in Collaborative Instruction Following ...
|
|
|
|
BASE
|
|
Show details
|
|
5 |
Learning Feature Weights using Reward Modeling for Denoising Parallel Corpora ...
|
|
|
|
BASE
|
|
Show details
|
|
6 |
Cross-lingual Aspect-based Sentiment Analysis with Aspect Term Code-Switching ...
|
|
|
|
BASE
|
|
Show details
|
|
7 |
Cross-lingual Transfer for Text Classification with Dictionary-based Heterogeneous Graph ...
|
|
|
|
BASE
|
|
Show details
|
|
8 |
NOAHQA: Numerical Reasoning with Interpretable Graph Question Answering Dataset ...
|
|
|
|
BASE
|
|
Show details
|
|
10 |
An Unsupervised Method for Building Sentence Simplification Corpora in Multiple Languages ...
|
|
|
|
BASE
|
|
Show details
|
|
11 |
Challenges in Detoxifying Language Models ...
|
|
|
|
Abstract:
Large language models (LM) generate remarkably fluent text and can be efficiently adapted across NLP tasks. Measuring and guaranteeing the quality of generated text in terms of safety is imperative for deploying LMs in the real world; to this end, prior work often relies on automatic evaluation of LM toxicity. We critically discuss this approach, evaluate several toxicity mitigation strategies with respect to both automatic and human evaluation, and analyze consequences of toxicity mitigation in terms of model bias and LM quality. We demonstrate that while basic intervention strategies can effectively optimize previously established automatic metrics on the REALTOXICITYPROMPTS dataset, this comes at the cost of reduced LM coverage for both texts about, and dialects of, marginalized groups. Additionally, we find that human raters often disagree with high automatic toxicity scores after strong toxicity reduction interventions—highlighting further the nuances involved in careful evaluation of LM toxicity. ...
|
|
URL: https://dx.doi.org/10.48448/y9w5-0x52 https://underline.io/lecture/38287-challenges-in-detoxifying-language-models
|
|
BASE
|
|
Hide details
|
|
12 |
SD-QA: Spoken Dialectal Question Answering for the Real World ...
|
|
|
|
BASE
|
|
Show details
|
|
13 |
Plan-then-Generate: Controlled Data-to-Text Generation via Planning ...
|
|
|
|
BASE
|
|
Show details
|
|
14 |
Sparsity and Sentence Structure in Encoder-Decoder Attention of Summarization Systems ...
|
|
|
|
BASE
|
|
Show details
|
|
15 |
Identity-Based Patterns in Deep Convolutional Networks: Generative Adversarial Phonology and Reduplication ...
|
|
|
|
BASE
|
|
Show details
|
|
16 |
Live Session - 4E: Phonology, Morphology and Word Segmentation ...
|
|
|
|
BASE
|
|
Show details
|
|
19 |
Rule-based Morphological Inflection Improves Neural Terminology Translation ...
|
|
|
|
BASE
|
|
Show details
|
|
20 |
Translating Headers of Tabular Data: A Pilot Study of Schema Translation ...
|
|
|
|
BASE
|
|
Show details
|
|
|
|