sci-dev

community

AI & ML interests

None defined yet.

fqjiang

authored a paper 5 months ago

SafeChain: Safety of Language Models with Long Chain-of-Thought Reasoning Capabilities

Paper • 2502.12025 • Published Feb 17 • 3

fqjiang

authored a paper 6 months ago

Small Models Struggle to Learn from Strong Reasoners

Paper • 2502.12143 • Published Feb 17 • 40

fqjiang

authored 2 papers 9 months ago

CleanGen: Mitigating Backdoor Attacks for Generation Tasks in Large Language Models

Paper • 2406.12257 • Published Jun 18, 2024

Stronger Models are NOT Stronger Teachers for Instruction Tuning

Paper • 2411.07133 • Published Nov 11, 2024 • 39

fqjiang

authored 5 papers about 1 year ago

ChatBug: A Common Vulnerability of Aligned LLMs Induced by Chat Templates

Paper • 2406.12935 • Published Jun 17, 2024 • 2

SafeDecoding: Defending against Jailbreak Attacks via Safety-Aware Decoding

Paper • 2402.08983 • Published Feb 14, 2024 • 5

ArtPrompt: ASCII Art-based Jailbreak Attacks against Aligned LLMs

Paper • 2402.11753 • Published Feb 19, 2024 • 6

BadChain: Backdoor Chain-of-Thought Prompting for Large Language Models

Paper • 2401.12242 • Published Jan 20, 2024 • 1

Magpie: Alignment Data Synthesis from Scratch by Prompting Aligned LLMs with Nothing

Paper • 2406.08464 • Published Jun 12, 2024 • 70