Tag
CRAFT is a unified counterfactual reasoning framework that improves tabular question answering and fact verification by constructing both original and counterfactual statements, extracting evidence from bidirectional reasoning paths, and integrating them via a weighted mechanism. Experiments show consistent improvements over baselines on WikiTQ and TabFact datasets.
Introduces Discrete-WAM, a unified discrete latent vision-action world policy that enables compositional causal reasoning and counterfactual reasoning in autonomous driving through aligned discrete tokens and a shared discrete diffusion framework.
CRONOS is a benchmark that evaluates counterfactual physical consistency in video prediction models by intervening on viewpoint, scene, object category, and appearance while keeping physical event types fixed. It reveals substantial failures in current video generators.
Introduces Implicit Behavior Policy Optimization (IBPO), a counterfactual comparison-based credit assignment framework that improves training stability and performance in multi-step reasoning tasks for large language models by converting sparse terminal rewards into step-sensitive learning signals.
该论文提出并评估了一类称为事件图基质的因果推理世界模型，通过确定性重放在类型化RDF事件日志上进行反事实查询，在多个基准上优于基线模型，同时保证了可检查性和可重放一致性。
This paper introduces NoisyCoconut, an inference-time method that improves LLM reliability by injecting noise into latent trajectories to generate diverse reasoning paths. The approach enables models to abstain when uncertain, significantly reducing error rates in mathematical reasoning tasks without requiring retraining.
This paper proposes a Three-in-One world model using Deep Boltzmann Machines for marketing intervention, combining energy-based consistency, outcome prediction, and counterfactual inference.