Released RL prompt injection attacker checkpoints from PIForge and PISmith.
Albert Yin
AlbertYin
AI & ML interests
LLM
Recent Activity
liked a model 1 day ago
AlbertYin/agentdojo_attacker_qwen3_4b_5.6_terra upvoted a paper 9 days ago
GraphForge: Training Working Agents with Graph-Anchored Workspace Synthesis upvoted a paper 10 days ago
PISmith: Reinforcement Learning-based Red Teaming for Prompt Injection DefensesOrganizations
None yet
models 9
AlbertYin/agentdojo_attacker_qwen3_4b_muse_spark
4B • Updated • 21
AlbertYin/agentdojo_attacker_qwen3_4b_5.6_terra_dev
4B • Updated • 10
AlbertYin/agentdojo_attacker_qwen3_4b_5.6_terra
4B • Updated • 28 • 1
AlbertYin/agentdojo_attacker_qwen3_4b_5.6_luna
4B • Updated • 20 • 1
AlbertYin/injecagent_attacker_qwen3_4b_secalign
4B • Updated • 15
AlbertYin/agentdojo_attacker_qwen3_4b_5_nano
4B • Updated • 39 • 1
AlbertYin/agentdojo_attacker_qwen3_4b_4o_mini
4B • Updated • 96
AlbertYin/piarena_attacker_qwen3_4b_secalign
4B • Updated • 27 • 2
AlbertYin/piarena_attacker_qwen3_4b_none
4B • Updated • 11