Built with and by Teycir Ben Soltane•
How to Use•FAQ•GitHub•arXiv.org•
Share:
ArXivCSExplorer
☆☆Bookmarks🏆RSSHow to UseFAQ
Home/Authors/Wendi Li

Wendi Li

4 indexed papers

Recent (6 mo)
4
With code
0
Influential cites
0
Benchmarked
0

Publications per year

4
26

Top categories

AI×4Robotics×2NLP×1Multiagent×1ML×1

Frequent co-authors

Sharon Li4×
Changdae Oh3×
Seongheon Park3×
Samuel Yeh3×
Zsolt Kira2×
Michael Hagenow2×

Research Timeline

2026
Hide-and-Seek in Trajectories: Discovering Failure Signals for VLA Runtime Monitoring

The paper proposes Hide-and-Seek, a novel framework that localizes failure signals in VLA model execution by treating failure detection as a coarsely supervised learning problem using contrastive objectives.

Hide-and-Seek in Trajectories: Discovering Failure Signals for VLA Runtime Monitoring

The paper proposes Hide-and-Seek, a novel framework that localizes failure signals in VLA model execution by treating failure detection as a coarsely supervised learning problem using contrastive objectives.

Neglected Free Lunch from Post-training: Progress Advantage for LLM Agents

The paper shows that reinforcement learning post-training can be used to create effective step-level scoring without dedicated reward model training.

Agentic World Modeling for Agent Test-time Scaling without Environment Replica

This paper introduces Agentic World Modeling (AWM), a method for safe and scalable agentic test-time scaling in shared, non-resettable environments.

Highlighted terms show continued research focus across papers

Papers

cs.AIcs.CLcs.MAEmpiricalRecentOct 29, 2026

Agentic World Modeling for Agent Test-time Scaling without Environment Replica

Wendi Li, Huseyin A. Inan, Wei-Ning Chen, Yanzhi Chen +3 more

This paper introduces Agentic World Modeling (AWM), a method for safe and scalable agentic test-time scaling in shared, non-resettable environments.

View →
cs.LGcs.AIEmpirical
Recent
Jun 24, 2026

Neglected Free Lunch from Post-training: Progress Advantage for LLM Agents

Changdae Oh, Wendi Li, Seongheon Park, Samuel Yeh +2 more

The paper shows that reinforcement learning post-training can be used to create effective step-level scoring without dedicated reward model training.

View →
cs.ROcs.AIRecentMay 29, 2026

Hide-and-Seek in Trajectories: Discovering Failure Signals for VLA Runtime Monitoring

Seongheon Park, Wendi Li, Changdae Oh, Samuel Yeh +3 more

The paper proposes Hide-and-Seek, a novel framework that localizes failure signals in VLA model execution by treating failure detection as a coarsely supervised learning problem using contrastive obje…

View →
cs.ROcs.AIRecentMay 29, 2026

Hide-and-Seek in Trajectories: Discovering Failure Signals for VLA Runtime Monitoring

Seongheon Park, Wendi Li, Changdae Oh, Samuel Yeh +3 more

The paper proposes Hide-and-Seek, a novel framework that localizes failure signals in VLA model execution by treating failure detection as a coarsely supervised learning problem using contrastive obje…

View →