21 citations · 46 across the 25 of their papers we have counts for
26 papers
Mechanist: AI as a Scientific Instrument for Discovering the Mechanisms of Intelligence
Mengru Wang, Junfeng Fang, Shuofei Qiao +17
AI models are increasingly used in scientific discovery and human decision-making. Yet how AI models work and what risks they pose remain poorly understood. As AI development becom…
PRISM Edit: One Vector for All Temporal Answers
Chen Huang, Qi Zheng, Ruiqin Zheng +2
Model editing keeps large language models (LLMs) up to date without retraining, but temporal facts expose a limitation of the prevailing locate-and-edit paradigm: an update is not…
MemTrapBench: Benchmarking Cognitive Traps in LLM Memory Use
Mengru Wang, Haozhe Luo, Zhenqian Xu +6
Memory has become a key component of large language models, enabling them to retain information and learn from long-term interactions. However, existing memory benchmarks mainly ev…
MobileMem: Learning from a Year of Mobile Experiences
Xinle Deng, Yida Xue, Xiangyuan Ru +16
The next generation of AI agents is increasingly moving beyond systems that answer isolated questions toward persistent personal assistants that can understand, remember, and conti…
Nemotron 3 Ultra: Open, Efficient Mixture-of-Experts Hybrid Mamba-Transformer Model for Agentic Reasoning
NVIDIA, :, Aaron Blakeman +572
We introduce Nemotron 3 Ultra, a 550 billion total and 55 billion active parameter Mixture-of-Experts Hybrid Mamba-Attention language model. We pre-trained Nemotron 3 Ultra on 20 t…
When Should Models Change Their Minds? Contextual Belief Management in Large Language Models
Haoming Xu, Weihong Xu, Zongrui Li +6
Long-horizon interactions require language models to manage accumulating information: when to update their state, when to preserve their state, and what to ignore. We study this ch…
SciAtlas: A Computable Atlas of Science for Knowledge-Grounded AI Research
Shuofei Qiao, Yunxiang Wei, Busheng Zhang +16
Artificial intelligence is rapidly entering the core workflows of scientific research. Yet reliable scientific reasoning requires access to accumulated scientific knowledge with su…
SkillNet: Create, Evaluate, and Connect AI Skills
Yuan Liang, Ruobin Zhong, Haoming Xu +47
Current AI agents can flexibly invoke tools and execute complex tasks, yet their long-term advancement is hindered by the lack of systematic accumulation and transfer of skills. Wi…
From Data to Behavior: Predicting Unintended Model Behaviors Before Training
Mengru Wang, Zhenqian Xu, Junfeng Fang +4
Large Language Models (LLMs) can acquire unintended biases from seemingly benign training data even without explicit cues or malicious content. Existing methods struggle to detect…
Why Steering Works: Toward a Unified View of Language Model Parameter Dynamics
Ziwen Xu, Chenyan Wu, Hengyu Sun +9
Methods for controlling large language models (LLMs), including local weight fine-tuning, LoRA-based adaptation, and activation-based interventions, are often studied in isolation,…
LightMem: Lightweight and Efficient Memory-Augmented Generation
Jizhan Fang, Xinle Deng, Haoming Xu +9
Despite their remarkable capabilities, Large Language Models (LLMs) struggle to effectively leverage historical interaction information in dynamic and complex environments. Memory…
OceanGym: A Benchmark Environment for Underwater Embodied Agents
Yida Xue, Mingjun Mao, Xiangyuan Ru +9
We introduce OceanGym, the first comprehensive benchmark for ocean underwater embodied agents, designed to advance AI in one of the most demanding real-world environments. Unlike t…
Automating Steering for Safe Multimodal Large Language Models
Lyucheng Wu, Mengru Wang, Ziwen Xu +4
Recent progress in Multimodal Large Language Models (MLLMs) has unlocked powerful cross-modal reasoning abilities, but also raised new safety concerns, particularly when faced with…
Beyond Prompt Engineering: Robust Behavior Control in LLMs via Steering Target Atoms
Mengru Wang, Ziwen Xu, Shengyu Mao +4
Precise control over language model generation is vital for ensuring both safety and reliability. Although prompt engineering and steering are commonly used to intervene in model b…
Two Experts Are All You Need for Steering Thinking: Reinforcing Cognitive Effort in MoE Reasoning Models Without Additional Training
Mengru Wang, Xingyu Chen, Yue Wang +12
Mixture-of-Experts (MoE) architectures within Large Reasoning Models (LRMs) have achieved impressive reasoning capabilities by selectively activating experts to facilitate structur…
EasyEdit2: An Easy-to-use Steering Framework for Editing Large Language Models
Ziwen Xu, Shuxun Wang, Kewei Xu +7
In this paper, we introduce EasyEdit2, a framework designed to enable plug-and-play adjustability for controlling Large Language Model (LLM) behaviors. EasyEdit2 supports a wide ra…
LookAhead Tuning: Safer Language Models via Partial Answer Previews
Kangwei Liu, Mengru Wang, Yujie Luo +7
Fine-tuning enables large language models (LLMs) to adapt to specific domains, but often compromises their previously established safety alignment. To mitigate the degradation of m…
ReLearn: Unlearning via Learning for Large Language Models
Haoming Xu, Ningyuan Zhao, Liming Yang +7
Current unlearning methods for large language models usually rely on reverse optimization to reduce target token probabilities. However, this paradigm disrupts the subsequent token…
Knowledge Mechanisms in Large Language Models: A Survey and Perspective
Mengru Wang, Yunzhi Yao, Ziwen Xu +10
Understanding knowledge mechanisms in Large Language Models (LLMs) is crucial for advancing towards trustworthy AGI. This paper reviews knowledge mechanism analysis from a novel ta…
To Forget or Not? Towards Practical Knowledge Unlearning for Large Language Models
Bozhong Tian, Xiaozhuan Liang, Siyuan Cheng +6
Large Language Models (LLMs) trained on extensive corpora inevitably retain sensitive data, such as personal privacy information and copyrighted material. Recent advancements in kn…
Knowledge Circuits in Pretrained Transformers
Yunzhi Yao, Ningyu Zhang, Zekun Xi +4
The remarkable capabilities of modern large language models are rooted in their vast repositories of knowledge encoded within their parameters, enabling them to perceive the world…
Detoxifying Large Language Models via Knowledge Editing
Mengru Wang, Ningyu Zhang, Ziwen Xu +7
This paper investigates using knowledge editing techniques to detoxify Large Language Models (LLMs). We construct a benchmark, SafeEdit, which covers nine unsafe categories with va…
A Comprehensive Study of Knowledge Editing for Large Language Models
Ningyu Zhang, Yunzhi Yao, Bozhong Tian +19
Large Language Models (LLMs) have shown extraordinary capabilities in understanding and generating text that closely mirrors human communication. However, a primary limitation lies…
Editing Personality for Large Language Models
Shengyu Mao, Xiaohan Wang, Mengru Wang +4
This paper introduces an innovative task focused on editing the personality traits of Large Language Models (LLMs). This task seeks to adjust the models' responses to opinion-relat…