HealthCraft is the first public RL safety environment for emergency medicine that evaluates frontier LLMs on trajectory-level safety with a dual-layer rubric, showing low multi-step performance and high safety failure rates.
Title resolution pending
5 Pith papers cite this work. Polarity classification is still indexing.
5
Pith papers citing it
citation-role summary
background 1
citation-polarity summary
years
2026 5roles
background 1polarities
background 1representative citing papers
Pore-scale DNS shows intermittency as a network-coupled process of drainage-imbibition cycles that enhances overall fluid mobility and produces the sub-linear macroscopic scaling regime.
Three Metapath2Vec variants create ingredient embeddings by walking a co-occurrence graph from recipes, a typed chemical compound graph from FlavorDB, or a controlled blend of both.
ConvLSTM trained self-supervised on simulated daily all-sky maps detects transients in Fermi-LAT data via pixel-wise residual anomalies with spatial filtering.
citing papers explorer
-
HealthCraft: A Reinforcement Learning Safety Environment for Emergency Medicine
HealthCraft is the first public RL safety environment for emergency medicine that evaluates frontier LLMs on trajectory-level safety with a dual-layer rubric, showing low multi-step performance and high safety failure rates.
-
Intermittent two-phase flow in porous media: insights from pore-scale direct numerical simulation
Pore-scale DNS shows intermittency as a network-coupled process of drainage-imbibition cycles that enhances overall fluid mobility and produces the sub-linear macroscopic scaling regime.
-
Epicure: Navigating the Emergent Geometry of Food Ingredient Embeddings
Three Metapath2Vec variants create ingredient embeddings by walking a co-occurrence graph from recipes, a typed chemical compound graph from FlavorDB, or a controlled blend of both.
-
Self-Supervised ConvLSTM for Fermi Large Area Telescope Transient Detection
ConvLSTM trained self-supervised on simulated daily all-sky maps detects transients in Fermi-LAT data via pixel-wise residual anomalies with spatial filtering.
- ScenePilot: Controllable Boundary-Driven Critical Scenario Generation for Autonomous Driving