Standard MORL metrics do not measure whether preference inputs reliably control agent behavior, so a new controllability metric is introduced to restore the link between user intent and agent output.
de las Heras Molins, E
4 Pith papers cite this work. Polarity classification is still indexing.
4
Pith papers citing it
abstract
Let us denote p(x|K) the space density of the points where identical particles of some kind, e.g. pi+ mesons, with momentum K are produced. When using the HBT method to determine p(x|K) one encounters ambiguities. We show that these ambiguities do not affect the even cumulants of the distribution p(x|K). In particular, the HBT radii of the homogeneity regions, which are given by the second order cumulants, and the distribution of distances between the pairs of production points for particles with momentum K can be reliably measured. The odd cumulants are ambiguous. The are, however, correlated. In particular, when the average position <x>(K) is known as a function of K there is no further ambiguity.
years
2026 4verdicts
UNVERDICTED 4representative citing papers
FAT decomposes structured prediction into specialist hypothesis generation and foundation-model proxy reasoning, yielding consistent gains over baselines on detection, trajectory, and segmentation tasks.
Response maps from game theory are used to model and differentiate human driver acceleration behaviors in response to yielding, non-yielding, and responsive AV actions based on simulator experiments.
Presents a scalable agent-based simulation framework for NYC transit fare policies that finds modest ridership effects but large shifts in mode use and income-group impacts, with fare-free buses aiding lower-income riders at revenue cost.
citing papers explorer
-
Controllability in preference-conditioned multi-objective reinforcement learning
Standard MORL metrics do not measure whether preference inputs reliably control agent behavior, so a new controllability metric is introduced to restore the link between user intent and agent output.
-
Rethinking Foundation Model Collaboration: Enhancing Specialized Models through Proxy Task Reasoning
FAT decomposes structured prediction into specialist hypothesis generation and foundation-model proxy reasoning, yielding consistent gains over baselines on detection, trajectory, and segmentation tasks.
-
Characterizing Driver Interactions with Autonomous Vehicles via Response Maps
Response maps from game theory are used to model and differentiate human driver acceleration behaviors in response to yielding, non-yielding, and responsive AV actions based on simulator experiments.
-
Simulating Public Transit Fare Policies in NYC: An Efficient, Socioeconomic-Aware Framework
Presents a scalable agent-based simulation framework for NYC transit fare policies that finds modest ridership effects but large shifts in mode use and income-group impacts, with fare-free buses aiding lower-income riders at revenue cost.