Mengru Wang
- Papers
- 15
Cite
Notes
Only stored in your browser.
Authored papers
15SciAtlas: A Large-Scale Knowledge Graph for Automated Scientific Research
arXiv 2026
SkillNet: Create, Evaluate, and Connect AI Skills
arXiv 2026
From Data to Behavior: Predicting Unintended Model Behaviors Before Training
arXiv 2026
Why Steering Works: Toward a Unified View of Language Model Parameter Dynamics
arXiv 2026
Nemotron 3 Nano: Open, Efficient Mixture-of-Experts Hybrid Mamba-Transformer Model for Agentic Reasoning
arXiv 2025
EasyEdit2: An Easy-to-use Steering Framework for Editing Large Language Models
arXiv 2025
Beyond Prompt Engineering: Robust Behavior Control in LLMs via Steering Target Atoms
arXiv 2025
LightMem: Lightweight and Efficient Memory-Augmented Generation
arXiv 2025
OceanGym: A Benchmark Environment for Underwater Embodied Agents
arXiv 2025
Automating Steering for Safe Multimodal Large Language Models
arXiv 2025
Too Good to be Bad: On the Failure of LLMs to Role-Play Villains
arXiv 2025
LookAhead Tuning: Safer Language Models via Partial Answer Previews
arXiv 2025
Knowledge Circuits in Pretrained Transformers
arXiv 2024
To Forget or Not? Towards Practical Knowledge Unlearning for Large Language Models
arXiv 2024
Unveiling the Pitfalls of Knowledge Editing for Large Language Models
arXiv 2023
Affiliations
Frequent co-authors
10from 15 papers