LLM-Tuning-Safety

LLM-Tuning-Safety's activity

• All
• Models
• Datasets
• Spaces
• Papers
• Collections
• Community
• Posts

vtu81 

authored a paper 7 months ago

SORRY-Bench: Systematically Evaluating Large Language Model Safety Refusal Behaviors

Paper • 2406.14598 • Published Jun 20, 2024

vtu81 

authored 3 papers 8 months ago

Assessing the Brittleness of Safety Alignment via Pruning and Low-Rank Modifications

Paper • 2402.05162 • Published Feb 7, 2024 • 1

BaDExpert: Extracting Backdoor Functionality for Accurate Backdoor Input Detection

Paper • 2308.12439 • Published Aug 23, 2023

Fine-tuning Aligned Language Models Compromises Safety, Even When Users Do Not Intend To!

Paper • 2310.03693 • Published Oct 5, 2023 • 1

yizeng 

authored 3 papers 10 months ago

Introducing v0.5 of the AI Safety Benchmark from MLCommons

Paper • 2404.12241 • Published Apr 18, 2024 • 11

RigorLLM: Resilient Guardrails for Large Language Models against Undesired Content

Paper • 2403.13031 • Published Mar 19, 2024 • 1

Fine-tuning Aligned Language Models Compromises Safety, Even When Users Do Not Intend To!

Paper • 2310.03693 • Published Oct 5, 2023 • 1

breakend 

authored a paper 11 months ago

On the Societal Impact of Open Foundation Models

Paper • 2403.07918 • Published Feb 27, 2024 • 17

breakend 

authored a paper over 1 year ago

Foundation Models and Fair Use

Paper • 2303.15715 • Published Mar 28, 2023 • 2

Unispac 

authored a paper over 1 year ago

Uncovering Adversarial Risks of Test-Time Adaptation

Paper • 2301.12576 • Published Jan 29, 2023

breakend 

authored a paper over 1 year ago

Cheaply Evaluating Inference Efficiency Metrics for Autoregressive Transformer APIs

Paper • 2305.02440 • Published May 3, 2023 • 1

breakend 

authored 2 papers almost 2 years ago

Pile of Law: Learning Responsible Data Filtering from the Law and a 256GB Open-Source Legal Dataset

Paper • 2207.00220 • Published Jul 1, 2022 • 3

When Does Pretraining Help? Assessing Self-Supervised Learning for Law and the CaseHOLD Dataset

Paper • 2104.08671 • Published Apr 18, 2021