Dialect-aware LLM safety classifiers from the DIA-GUARD project — fine-tuned on 48 English dialects for safe/unsafe content detection.
Jason Lucas
jsl5710
·
AI & ML interests
Trustworthy AI, Multilingual NLP, Low-Resource Languages, Safe AI, Transfer Learning, AI for Cybersecurity, Human-Center AI, Privacy and Security
Recent Activity
upvoted a paper 2 days ago
Moderation by Interrogation: Uncovering Concealed Intent in Agent Social Networks updated a model 2 months ago
jsl5710/TruthTorchLM-HC-proxies published a model 2 months ago
jsl5710/TruthTorchLM-HC-proxiesOrganizations
models 68
jsl5710/TruthTorchLM-HC-proxies
Updated
jsl5710/Shield-Qwen3Guard-Gen-0.6B-KD-TED-Qwen3Guard-Gen-8B-FT-aliwister-S2-QG-PEFT
0.6B • Updated • 7
jsl5710/Shield-Qwen3Guard-Gen-0.6B-KD-MINILLM-Qwen3Guard-Gen-8B-FT-aliwister-S2-QG-PEFT
0.6B • Updated • 4
jsl5710/Shield-Qwen3-1.7B-KD-MINILLM-Qwen3Guard-Gen-8B-FT-aliwister-S2-Q17-PEFT
2B • Updated • 6
jsl5710/Shield-Qwen3Guard-Gen-0.6B-KD-GKD-Qwen3Guard-Gen-8B-FT-aliwister-S2-QG-PEFT
0.6B • Updated • 7
jsl5710/Shield-Qwen3-1.7B-KD-GKD-Qwen3Guard-Gen-8B-FT-aliwister-S2-Q17-PEFT
2B • Updated • 4
jsl5710/Shield-Qwen3Guard-Gen-8B-FT-PEFT-CE
8B • Updated • 5
jsl5710/Shield-Qwen3-1.7B-KD-TED-Qwen3Guard-Gen-8B-FT-S2-Q17-BASE
2B • Updated • 5
jsl5710/Shield-Qwen3Guard-Gen-0.6B-KD-TED-Qwen3Guard-Gen-8B-FT-S2-QG-BASE
0.6B • Updated • 7
jsl5710/Shield-Qwen3-1.7B-KD-TED-Qwen3Guard-Gen-8B-FT-S2-Q17-FULL
2B • Updated • 10