Picture for Melanie Subbiah

Melanie Subbiah

STORYSUMM: Evaluating Faithfulness in Story Summarization

Add code
Jul 09, 2024
Viaarxiv icon

Reading Subtext: Evaluating Large Language Models on Short Story Summarization with Writers

Add code
Mar 02, 2024
Viaarxiv icon

Check-COVID: Fact-Checking COVID-19 News Claims with Scientific Evidence

Add code
May 29, 2023
Viaarxiv icon

Unsupervised Selective Rationalization with Noise Injection

Add code
May 27, 2023
Viaarxiv icon

Detecting Harmful Agendas in News Articles

Add code
Jan 31, 2023
Viaarxiv icon

SafeText: A Benchmark for Exploring Physical Safety in Language Models

Add code
Oct 18, 2022
Figure 1 for SafeText: A Benchmark for Exploring Physical Safety in Language Models
Figure 2 for SafeText: A Benchmark for Exploring Physical Safety in Language Models
Figure 3 for SafeText: A Benchmark for Exploring Physical Safety in Language Models
Figure 4 for SafeText: A Benchmark for Exploring Physical Safety in Language Models
Viaarxiv icon

Mitigating Covertly Unsafe Text within Natural Language Systems

Add code
Oct 17, 2022
Figure 1 for Mitigating Covertly Unsafe Text within Natural Language Systems
Figure 2 for Mitigating Covertly Unsafe Text within Natural Language Systems
Figure 3 for Mitigating Covertly Unsafe Text within Natural Language Systems
Figure 4 for Mitigating Covertly Unsafe Text within Natural Language Systems
Viaarxiv icon

Language Models are Few-Shot Learners

Add code
Jun 05, 2020
Figure 1 for Language Models are Few-Shot Learners
Figure 2 for Language Models are Few-Shot Learners
Figure 3 for Language Models are Few-Shot Learners
Figure 4 for Language Models are Few-Shot Learners
Viaarxiv icon