arxiv:2601.23143
Sangwoo Park
Sangsang
AI & ML interests
I do LLM Safety & Reasoning research (KAIST AI)
Recent Activity
updated a dataset about 9 hours ago
Sangsang/Qwen3-8B-thinksafe-unfiltered-raw published a dataset about 9 hours ago
Sangsang/Qwen3-8B-thinksafe-unfiltered-raw updated a model about 11 hours ago
Sangsang/DeepSeek-R1-Distill-Llama-8B_from_Distill-Qwen-32BOrganizations
None yet
models 180
Sangsang/DeepSeek-R1-Distill-Llama-8B_from_Distill-Qwen-32B
Text Generation • Updated
Sangsang/qwen3-8B-thinksafe-Qwen-8B-unfiltered-32-pm-3ep
Text Generation • Updated • 12
Sangsang/R1-8B-thinksafe-DeepSeek-8B-unfiltered-32-pm-3ep
Text Generation • Updated • 12
Sangsang/qwen3-14B-safechain-32-pm-3ep
Text Generation • Updated • 22
Sangsang/R1-14B-safechain-32-pm-3ep
Text Generation • Updated • 11
Sangsang/feedback_disallowed_ema_DeepSeek-R1-Distill-Llama-8B_reverse_kl_ema0p999_ep30
Text Generation • Updated • 20
Sangsang/grpo_DeepSeek-R1-Distill-Llama-8B_bs8_g16_mb128_lr1e-6_b1e-3_clip0p2_temp0p7_ep30
Text Generation • Updated • 12
Sangsang/R1-14B-thinksafe-DeepSeek-14B-32-pm-3ep
Text Generation • Updated • 18
Sangsang/qwen3-14B-thinksafe-Qwen-14B-32-pm-3ep
Text Generation • Updated • 19
Sangsang/ci-feedback_weighted_asym_bi_kl_fixed_ema_Qwen2.5-7B-Instruct_bw1p6_fw0p4_ema0p999_ep30
Text Generation • 8B • Updated • 24
datasets 94
Sangsang/Qwen3-8B-thinksafe-unfiltered-raw
Viewer • Updated • 40k
Sangsang/DeepSeek-R1-Distill-Llama-8B-thinksafe-unfiltered-raw
Viewer • Updated • 40k • 11
Sangsang/CI-DeepSeek-R1-Distill-Qwen-32B-Augmented-Responses
Viewer • Updated • 729 • 10
Sangsang/ContextualIntegritySyntheticDataset_DeepSeek-R1-Distill-Qwen-32B_all
Viewer • Updated • 729 • 7
Sangsang/ContextualIntegritySyntheticDataset_Olmo-3-7B-Think_all
Viewer • Updated • 729 • 27
Sangsang/ThinkSafe-R1-Distill-8B_unfiltered
Viewer • Updated • 40k • 7
Sangsang/ThinkSafe-Qwen3-8B_unfiltered
Preview • Updated • 7
Sangsang/DeepSeek-R1-Distill-Qwen-14B-thinksafe
Viewer • Updated • 39.8k • 19
Sangsang/CI_v2_Qwen3-14B_all
Viewer • Updated • 729 • 18
Sangsang/CI_v2_Qwen3-4B_all
Viewer • Updated • 729 • 23