arxiv:2605.17873
Woongyeong Yeo
wgcyeo
AI & ML interests
None yet
Recent Activity
authored a paper 1 day ago
HINT-SD: Targeted Hindsight Self-Distillation for Long-Horizon Agents submitted a paper 2 days ago
HINT-SD: Targeted Hindsight Self-Distillation for Long-Horizon Agents authored a paper 5 days ago
It Takes Two: Complementary Self-Distillation for Contextual Integrity in LLMsOrganizations
None yet
models 67
wgcyeo/ci-feedback_weighted_asym_bi_kl_fixed_ema_Qwen3-4B-Instruct-2507_bw0p5_fw0p5_ema0p9999_ep30
Text Generation • Updated • 11
wgcyeo/ci-feedback_weighted_asym_bi_kl_fixed_ema_Qwen3-4B-Instruct-2507_bw0p5_fw0p5_ema0p99_ep30
Text Generation • Updated • 17
wgcyeo/ci-feedback_weighted_asym_bi_kl_fixed_ema_Qwen3-4B_bw0p5_fw0p5_ema0p999_ep30_rgcomq0p5
Text Generation • Updated • 13
wgcyeo/ci-feedback_asym_bi_kl_hybrid_fixed_ema_Qwen3-14B_bw0p5_fw0p5_ema0p999_ep30
Text Generation • Updated • 20
wgcyeo/ci-feedback_asym_bi_kl_hybrid_fixed_ema_Qwen3-8B_bw0p5_fw0p5_ema0p999_ep30
Text Generation • Updated • 13
wgcyeo/ci-sft_Qwen3-4B-Instruct-2507_lr1e-6_ep30_self
Text Generation • Updated • 12
wgcyeo/ci-grpo_Olmo-3-7B-Think_bs8_g16_mb128_lr1e-6_b1e-3_clip0p2_temp0p7_ep30
Text Generation • Updated • 11
wgcyeo/ci-feedback_asym_bi_kl_hybrid_fixed_ema_Qwen3-1.7B_bw0p5_fw0p5_ema0p999_ep30
Text Generation • Updated • 16
wgcyeo/feedback_asym_bi_kl_hybrid_fixed_ema_Qwen3-0.6B_bw0p5_fw0p5_ema0p999_ep30
Text Generation • Updated • 14
wgcyeo/ci-feedback_weighted_asym_bi_kl_fixed_interp_Qwen3-4B-Instruct-2507_bw0p5_fw0p5_stw0p3_ep30
Text Generation • Updated • 7
datasets 29
wgcyeo/CI-Qwen2.5-32B-Instruct-Augmented-Responses
Viewer • Updated • 729 • 19
wgcyeo/ContextualIntegritySyntheticDataset_Qwen3-14B_all
Viewer • Updated • 729 • 80 • 1
wgcyeo/ContextualIntegritySyntheticDataset_Qwen3-8B_all
Viewer • Updated • 729 • 53
wgcyeo/ContextualIntegritySyntheticDataset_Qwen3-1.7B_all
Viewer • Updated • 729 • 70
wgcyeo/ContextualIntegritySyntheticDataset_Qwen3-0.6B_all
Viewer • Updated • 729 • 17
wgcyeo/CI-Qwen3-32B-THINK-OFF-Augmented-Responses
Viewer • Updated • 729 • 110
wgcyeo/CI-Qwen3-4B-Instruct-2507-Augmented-Responses
Viewer • Updated • 729 • 17
wgcyeo/CI-Qwen3-4B-Augmented-Responses
Viewer • Updated • 729 • 70
wgcyeo/CI-DeepSeek-R1-Distill-Llama-70B-Augmented-Responses
Viewer • Updated • 729 • 46
wgcyeo/ContextualIntegritySyntheticDataset_DeepSeek-R1-Distill-Llama-70B_all
Viewer • Updated • 729 • 12