https://arxiv.org/pdf/2601.18081
韩沛煊
HakHan
·
AI & ML interests
None yet
Recent Activity
authored
a paper
about 19 hours ago
SafeSwitch: Steering Unsafe LLM Behavior via Internal Activation Signals
authored
a paper
about 19 hours ago
Energy-Based Transformers are Scalable Learners and Thinkers
authored
a paper
about 19 hours ago
EscapeBench: Towards Advancing Creative Intelligence of Language Model
Agents
Organizations
models
9
HakHan/DRPG_Planner
Updated
HakHan/DRPG_JudgeModel
4B
•
Updated
•
1
HakHan/SafeSwitch
Updated
HakHan/Qwen2.5-3B-Instruct-Persuader
3B
•
Updated
•
2
HakHan/Qwen2.5-3B-Instruct-ToMAP
3B
•
Updated
•
1
HakHan/ToMAP-Predictor
Updated
HakHan/gpt-neox-20b-single-digit
Updated
HakHan/Web-Graph-Embedding
Feature Extraction
•
Updated
•
4
HakHan/Web-DRO-DR
Feature Extraction
•
Updated
•
1