Ideas in Motion

Representation Learning Enables Scalable Multitask Deep Reinforcement Learning

Local Guidance, Global Impact: Gaussian-Reshaped Trust Region Unlocks Behavior Transitions

Stable Deep Reinforcement Learning via Isotropic Gaussian Representations

Simplicial Embeddings Improve Sample Efficiency in Actor-Critic Agents

Asymmetric Proximal Policy Optimization: mini-critics boost LLM reasoning

Grounding Computer Use Agents on Human Demonstrations

Tricks or Traps? A deep dive into RL for LLM reasoning

Kaleidoscope: In-language Exams for Massively Multilingual Vision Evaluation

A mechanistic analysis of looped reasoning language models

Do Enterprise Systems Need Learned World Models? The Importance of Context to Infer Dynamics

Stable Gradients for Stable Learning at Scale in Deep Reinforcement Learning

The Courage to Stop: Overcoming Sunk Cost Fallacy in Deep Reinforcement Learning

The Impact of On-Policy Parallelized Data Collection on Deep Reinforcement Learning Networks

Mitigating Plasticity Loss in Continual Reinforcement Learning by Reducing Churn

Measure gradients, not activations! Enhancing neuronal activity in deep reinforcement learning

Generating Creative Chess Puzzles

Evaluating In Silico Creativity: An Expert Review of AI Chess Compositions

Trajectory balance with asynchrony: Decoupling exploration and learning for fast, scalable LLM post-training

Include: Evaluating multilingual language understanding with regional knowledge

All languages matter: Evaluating LMMs on culturally diverse 100 languages

Neuroplastic Expansion in Deep Reinforcement Learning

Don't flatten, tokenize! Unlocking the key to SoftMoE's efficacy in deep RL

Adaptive Computation Pruning for the Forgetting Transformer

Learning structured spatiotemporal tasks with xLSTM under uncertainty: A multi-task approach

Recursive self-aggregation unlocks deep thinking in large language models

A Comedy of Estimators: On KL Regularization in RL Training of LLMs

Mixtures of Experts Unlock Parameter Scaling for Deep RL

In value-based deep RL, a pruned network is a good network

On the consistency of hyper-parameter selection in value-based deep RL

JaxPruner: A concise library for sparsity research

Mixture of Experts in a Mixture of RL settings

Small batch deep reinforcement learning

Bigger, Better, Faster: Human-level Atari with human-level efficiency

Probabilistic multi-modal depth estimation based on camera–LiDAR sensor fusion

Revisiting Rainbow: Promoting more insightful and inclusive deep RL research

Lifting the veil on hyper-parameters for value-based deep reinforcement learning

Quantification of operating reserves with high penetration of wind power considering extreme values

Exploiting the potential of deep RL for classification tasks in high-dimensional and unstructured data

Probabilistic Perception System for Object Classification Based on Camera-LiDAR Sensor Fusion

An integrated OPF dispatching model with wind power and demand response for day-ahead markets

Divide and conquer: An accurate machine learning algorithm to process split videos on a parallel processing infrastructure

Evaluación del Rendimiento de Modulos Solares Híbridos (FV/T) Para el Abastecimiento Energético de Autoclaves Hospitalarias

Impacts of demand response under wind power uncertainty in network-constrained electricity markets

Quantification of operating reserves with wind power in day-ahead dispatching

Network Topological Notions for Power Systems Security Assessment