Action Semantics Learning trains app agents to align with the semantic effects of actions via a Semantic Estimator module, improving robustness to out-of-distribution scenarios over syntax-matching fine-tuning.
Understanding world or predicting future? a comprehensive survey of world models
5 Pith papers cite this work, alongside 2 external citations. Polarity classification is still indexing.
5
Pith papers citing it
2
external citations · external index
citation-role summary
background 2
citation-polarity summary
verdicts
UNVERDICTED 5roles
background 2polarities
background 2representative citing papers
World models enable efficient AI planning but create risks from adversarial corruption, goal misgeneralization, and human bias, demonstrated via attacks that amplify errors and reduce rewards on models like RSSM and DreamerV3.
Authors define EASE as a modular architecture for LLM multi-agent simulations, implement it in the SiliSocS sandbox, and illustrate its use via three case studies on research questions in generated social scenarios.
LLMs are reframed as a degenerate case of world models with a continuous spectrum of architectures from next-token prediction to joint-embedding predictive architectures.
Active inference offers a variational way to phenotype agency in AI systems by measuring empowerment in generative models via a T-maze paradigm.
citing papers explorer
-
Beyond Syntax: Action Semantics Learning for App Agents
Action Semantics Learning trains app agents to align with the semantic effects of actions via a Semantic Estimator module, improving robustness to out-of-distribution scenarios over syntax-matching fine-tuning.
-
Safety, Security, and Cognitive Risks in World Models
World models enable efficient AI planning but create risks from adversarial corruption, goal misgeneralization, and human bias, demonstrated via attacks that amplify errors and reduce rewards on models like RSSM and DreamerV3.
-
EASE Configuration Facilitates A Reproducible Science of LLM Social Simulations
Authors define EASE as a modular architecture for LLM multi-agent simulations, implement it in the SiliSocS sandbox, and illustrate its use via three case studies on research questions in generated social scenarios.
-
From Tokens to States: LLMs as a Special Case of World Models and the Continuous Path Beyond
LLMs are reframed as a degenerate case of world models with a continuous spectrum of architectures from next-token prediction to joint-embedding predictive architectures.
-
Active Inference: A method for Phenotyping Agency in AI systems?
Active inference offers a variational way to phenotype agency in AI systems by measuring empowerment in generative models via a T-maze paradigm.