Related papers: Entropy Regularized Belief Reporting
Although pretrained language models (PTLMs) contain significant amounts of world knowledge, they can still produce inconsistent answers to questions when probed, even after specialized training. As a result, it can be hard to identify what…
Reinforcement Learning with Verifiable Rewards (RLVR) has emerged as an important paradigm for unlocking reasoning capabilities in large language models, exemplified by the success of OpenAI o1 and DeepSeek-R1. Currently, Group Relative…
We examine how causal beliefs affect an agent's choices and how feedback on those choices leads to updated causal beliefs. Building on the structural-equations framework for modeling causality, we first examine the general problem of…
Stochastic and soft optimal policies resulting from entropy-regularized Markov decision processes (ER-MDP) are desirable for exploration and imitation learning applications. Motivated by the fact that such policies are sensitive with…
We study an opinion formation model by the means of a co-evolving complex network where the vertices represent the individuals, characterised by their evolving opinions, and the edges represent the interactions among them. The network…
Opinion formation models typically represent each individual as a single variable. However, in practice each individual holds interconnected beliefs whose internal organization may influence collective outcomes. To explore this dependence,…
Safety-critical autonomy in adversarial settings demands more than Lyapunov stability of tracking error signals. An agent executing a goal-directed trajectory is intrinsically legible to a passive observer running online Bayesian inference,…
Large language models (LLMs) are increasingly used as simulated participants in social science experiments, but their behavior is often unstable and highly sensitive to design choices. Prior evaluations frequently conflate base-model…
Autonomous agents powered by LLMs and Retrieval-Augmented Generation (RAG) are proficient consumers of digital content but remain unidirectional, a limitation we term epistemic asymmetry. This isolation leads to redundant reasoning and…
Interactive large language model (LLM) agents operating via multi-turn dialogue and multi-step tool calling are increasingly used in production. Benchmarks for these agents must both reliably compare models and yield on-policy training…
The BLP model is the workhorse framework in empirical IO and enables estimation of demand models for differentiated products using aggregate product shares. In practice, however, the share of the outside good is often unobserved. This paper…
We show how to train the fast dependency parser of Smith and Eisner (2008) for improved accuracy. This parser can consider higher-order interactions among edges while retaining O(n^3) runtime. It outputs the parse with maximum expected…
Entity-level unlearning is usually evaluated by what a model says: whether it stops naming the target, refuses a query, or shifts a Truth Ratio distribution. These output-level tests, however, do not show whether a subject's internal…
Mixture distributions are a workhorse model for multimodal data in information theory, signal processing, and machine learning. Yet even when each component density is simple, the differential entropy of the mixture is notoriously hard to…
Entropy regularization is used to get improved optimization performance in reinforcement learning tasks. A common form of regularization is to maximize policy entropy to avoid premature convergence and lead to more stochastic policies for…
In this contribution, we propose a generic online (also sometimes called adaptive or recursive) version of the Expectation-Maximisation (EM) algorithm applicable to latent variable models of independent observations. Compared to the…
We propose a model of unawareness that remains close to the paradigm of Aumann's model for knowledge [R. J. Aumann, International Journal of Game Theory 28 (1999) 263-300]: just as Aumann uses a correspondence on a state space to define an…
The method of Maximum (relative) Entropy (ME) is used to translate the information contained in the known form of the likelihood into a prior distribution for Bayesian inference. The argument is guided by intuition gained from the…
Preference-based reinforcement learning (PBRL) offers a promising alternative to explicit reward engineering by learning from pairwise trajectory comparisons. However, real-world preference data often comes from heterogeneous annotators…
Reinforcement learning algorithms typically rely on the assumption that the environment dynamics and value function can be expressed in terms of a Markovian state representation. However, when state information is only partially observable,…