2026 NeurIPS
Test-Time Risk Adaptation with Mixture of Agents
A test-time risk-adaptation method that composes source agents according to deployment-time risk specifications.
Research
01 / Research direction
Risk-Aware Learning and Reinforcement Learning
Methods for learning decisions under uncertainty, distribution shift, and explicit risk specifications.
This thread asks how policies should adapt when the cost of failure changes across environments, users, or operating regimes.
Papers
Related work in this direction
4 papers connected to this topic.
2025 ICML
LEVIS: Large Exact Verifiable Input Spaces for Neural Networks
A verification method for constructing large exact input spaces over which neural-network behavior can be certified.
2025 ICML
Bounded Rationality for LLMs: Satisficing Alignment at Inference-Time
An inference-time alignment method that treats acceptable behavior through satisficing rather than single-objective maximization.
2024 NAPS
Should We Use Model-Free or Model-Based Control? A Case Study of Battery Control
A case study comparing model-free and model-based control choices for battery control problems.