Copertina

Information-theoretic analysis of world models in optimal reward maximizers

Fuente: arXiv

Salvato in:

Dettagli Bibliografici
Autori principali:	Harwood, Alfred, Faustino, Jose, Altair, Alex
Natura:	Preprint
Pubblicazione:	2026
Soggetti:	Artificial Intelligence
Accesso online:	Acceder al recurso
Tags:	Aggiungi Tag Nessun Tag, puoi essere il primo ad aggiungerne!!

Documenti analoghi

What should be observed for optimal reward in POMDPs?
di: Konsta, Alyzia-Maria, et al.
Pubblicazione: (2024)

Agent-centric learning: from external reward maximization to internal knowledge curation
di: Zhou, Hanqi, et al.
Pubblicazione: (2025)

TripScore: Benchmarking and rewarding real-world travel planning with fine-grained evaluation
di: Qu, Yincen, et al.
Pubblicazione: (2025)

Towards better dense rewards in Reinforcement Learning Applications
di: Zhang, Shuyuan
Pubblicazione: (2025)

Reasoning over mathematical objects: on-policy reward modeling and test time aggregation
di: Aggarwal, Pranjal, et al.
Pubblicazione: (2026)

Active teacher selection for reward learning
di: Freedman, Rachel, et al.
Pubblicazione: (2023)

Self-rewarding correction for mathematical reasoning
di: Xiong, Wei, et al.
Pubblicazione: (2025)

Noise-based reward-modulated learning
di: Fernández, Jesús García, et al.
Pubblicazione: (2025)

Order-theoretic models for decision-making: Learning, optimization, complexity and computation
di: Hack, Pedro
Pubblicazione: (2024)

GDPO: Group reward-Decoupled Normalization Policy Optimization for Multi-reward RL Optimization
di: Liu, Shih-Yang, et al.
Pubblicazione: (2026)

Preference learning in shades of gray: Interpretable and bias-aware reward modeling for human preferences
di: Oprea, Simona-Vasilica, et al.
Pubblicazione: (2026)

Class-Conditional self-reward mechanism for improved Text-to-Image models
di: Ghazouali, Safouane El, et al.
Pubblicazione: (2024)

The impact of intrinsic rewards on exploration in Reinforcement Learning
di: Kayal, Aya, et al.
Pubblicazione: (2025)

EVAL: EigenVector-based Average-reward Learning
di: Adamczyk, Jacob, et al.
Pubblicazione: (2025)

Streaming Looking Ahead with Token-level Self-reward
di: Zhang, Hongming, et al.
Pubblicazione: (2025)

Episodic Reinforcement Learning with Expanded State-reward Space
di: Liang, Dayang, et al.
Pubblicazione: (2024)

Self-supervised network distillation: an effective approach to exploration in sparse reward environments
di: Pecháč, Matej, et al.
Pubblicazione: (2023)

R-ParVI: Particle-based variational inference through lens of rewards
di: Huang, Yongchao
Pubblicazione: (2025)

Deep Reinforcement Learning with anticipatory reward in LSTM for Collision Avoidance of Mobile Robots
di: Poulet, Olivier, et al.
Pubblicazione: (2025)

Bayesian Flow Networks
di: Graves, Alex, et al.
Pubblicazione: (2023)

HelpSteer2: Open-source dataset for training top-performing reward models
di: Wang, Zhilin, et al.
Pubblicazione: (2024)

Information-theoretic Distinctions Between Deception and Confusion
di: Young, Robin
Pubblicazione: (2025)

reward-lens: A Mechanistic Interpretability Library for Reward Models
di: Nadaf, Mohammed Suhail B
Pubblicazione: (2026)

MOSLIM:Align with diverse preferences in prompts through reward classification
di: Zhang, Yu, et al.
Pubblicazione: (2025)

Risk-averse Total-reward MDPs with ERM and EVaR
di: Su, Xihong, et al.
Pubblicazione: (2024)

Synthesizing world models for bilevel planning
di: Ahmed, Zergham, et al.
Pubblicazione: (2025)

Assessing model error in counterfactual worlds
di: Howerton, Emily, et al.
Pubblicazione: (2025)

LLM world models are mental: Output layer evidence of brittle world model use in LLM mechanical reasoning
di: Robertson, Cole, et al.
Pubblicazione: (2025)

Towards an Inferentialist Account of Information Through Proof-theoretic Semantics
di: Collinson, Matthew, et al.
Pubblicazione: (2026)

Just Say What You Want: Only-prompting Self-rewarding Online Preference Optimization
di: Xu, Ruijie, et al.
Pubblicazione: (2024)

Reasoning with maximal consistent signatures
di: Thimm, Matthias, et al.
Pubblicazione: (2024)

On the Weaknesses of Backdoor-based Model Watermarking: An Information-theoretic Perspective
di: Hu, Aoting, et al.
Pubblicazione: (2024)

Leveraging LLMs for reward function design in reinforcement learning control tasks
di: Cardenoso, Franklin, et al.
Pubblicazione: (2025)

Information-theoretic Bayesian Optimization: Survey and Tutorial
di: Garrido-Merchán, Eduardo C.
Pubblicazione: (2025)

BLEUBERI: BLEU is a surprisingly effective reward for instruction following
di: Chang, Yapei, et al.
Pubblicazione: (2025)

The geometry of invariant learning: an information-theoretic analysis of data augmentation and generalization
di: Bouyahia, Abdelali, et al.
Pubblicazione: (2026)

Zero-Incentive Dynamics: a look at reward sparsity through the lens of unrewarded subgoals
di: Molinghen, Yannick, et al.
Pubblicazione: (2025)

From monoliths to modules: Decomposing transducers for efficient world modelling
di: Boyd, Alexander, et al.
Pubblicazione: (2025)

A Pre-training Framework for Relational Data with Information-theoretic Principles
di: Truong, Quang, et al.
Pubblicazione: (2025)

On minimizers and convolutional filters: theoretical connections and applications to genome analysis
di: Yu, Yun William
Pubblicazione: (2021)