MiraBench defines action-conditioned reliability via three levels (physics adherence, action-following fidelity, optimism bias detection) and applies it to 12 model configurations using a 16,000-judgment human corpus, finding visual fidelity a poor proxy for action fidelity, no reliable scale benefi
hub
Temporal difference learning for model predictive control
23 Pith papers cite this work. Polarity classification is still indexing.
23
Pith papers citing it
hub tools
citation-role summary
background 3
citation-polarity summary
roles
background 3polarities
background 3representative citing papers
MSRL represents trajectory segments as PSD matrices to prove additive composition properties and bootstrap value functions for better transfer, reaching 0.73 AUC versus 0.57-0.65 baselines.
JEDI is the first online end-to-end latent diffusion world model that trains latents from denoising loss rather than reconstruction, achieving competitive Atari100k results with 43% less VRAM and over 3x faster sampling than pixel diffusion baselines.
RLA-WM predicts residual latent actions via flow matching to create visual feature world models that outperform prior feature-based and diffusion approaches while enabling offline video-based robot RL.
BiTrajDiff augments offline RL datasets by running independent forward and backward diffusion processes from intermediate states, yielding higher performance than prior one-directional data-augmentation baselines on D4RL.
Ada-Diffuser is a causal diffusion model that jointly learns observed interaction structure and underlying latent dynamics from minimal observations for adaptive planning and policy learning.
TRAP is a tail-aware ranking attack that plants a backdoor in world models so that a trigger causes the model to reorder a few critical imagined trajectories and redirect planning while preserving normal behavior on clean inputs.
RAY-TOLD combines ray-based latent dynamics from LiDAR with MPPI control and a learned policy prior via mixture sampling to lower collision rates in high-density dynamic obstacle environments compared to standard MPPI.
Single-stage fine-tuning of a video model to generate actions as latent frames plus future states and values yields state-of-the-art robot policy performance on LIBERO, RoboCasa, and bimanual tasks.
A controllable world model trained on the DROID dataset generates consistent multi-view robot trajectories for over 20 seconds and improves generalist policy success rates by 44.7% via imagined trajectory fine-tuning.
A delay-aware model-based RL framework with sequential belief filtering handles random out-of-sequence observations in POMDPs and outperforms MDP baselines while showing robustness to delay shifts.
EfficientTrack integrates model-based learning and closed-loop dynamics to minimize tracking errors in excavator trajectory control with high efficiency and precision, outperforming prior learning-based methods in simulation and real-world tests.
V-JEPA 2 pre-trained on massive unlabeled video achieves strong results on motion understanding and action anticipation, SOTA video QA at 8B scale, and enables zero-shot robotic planning on Franka arms using only 62 hours of unlabeled robot video.
Real-time chunking (RTC) allows diffusion- and flow-based action chunking policies to execute smoothly and asynchronously, maintaining high success rates on dynamic tasks even with significant inference latency.
DINO-WM builds world models on pre-trained DINOv2 features to enable zero-shot planning from offline data without rewards or demonstrations.
Valdi pairs a latent diffusion dynamics model with end-to-end MPC training and reports that one diffusion step matches an MLP baseline on CarRacing while exposing a multimodality-control trade-off.
MBDPO reformulates policy optimization as a diffusion process over searched trajectories in latent world models to reduce misalignment between search and value learning.
The paper presents stable-worldmodel (swm), a platform with high-performance data layer, modern world model baselines, planning solvers, and extended environments for reproducible research and generalization evaluation.
D2AC combines a diffusion actor with a distributional critic via fused distributional RL and clipped double Q-learning to reach state-of-the-art results on 18 hard control benchmarks including Humanoid, Dog, and Shadow Hand.
BYOL-γ uses self-predictive representations to approximate successor representations, improving zero-shot combinatorial generalization in goal-conditioned behavioral cloning.
SDPG is a new on-policy visual RL algorithm that estimates gradients via stochastic perturbations of rollouts, achieving faster training and lower memory use than baselines on visual MuJoCo tasks while adding new robotics benchmarks and sim-to-real results.
HiMAQ applies hierarchical vector quantization to human demonstrations to generate macro actions that yield higher human-likeness scores than flat MAQ on D4RL while matching or exceeding success rates across IQL, SAC, and RLPD.
citing papers explorer
-
MiraBench: Evaluating Action-Conditioned Reliability in Robotic World Models
MiraBench defines action-conditioned reliability via three levels (physics adherence, action-following fidelity, optimism bias detection) and applies it to 12 model configurations using a 16,000-judgment human corpus, finding visual fidelity a poor proxy for action fidelity, no reliable scale benefi
-
Matrix-Space Reinforcement Learning for Reusing Local Transition Geometry
MSRL represents trajectory segments as PSD matrices to prove additive composition properties and bootstrap value functions for better transfer, reaching 0.73 AUC versus 0.57-0.65 baselines.
-
JEDI: Joint Embedding Diffusion World Model for Online Model-Based Reinforcement Learning
JEDI is the first online end-to-end latent diffusion world model that trains latents from denoising loss rather than reconstruction, achieving competitive Atari100k results with 43% less VRAM and over 3x faster sampling than pixel diffusion baselines.
-
Learning Visual Feature-Based World Models via Residual Latent Action
RLA-WM predicts residual latent actions via flow matching to create visual feature world models that outperform prior feature-based and diffusion approaches while enabling offline video-based robot RL.
-
BiTrajDiff: Bidirectional Trajectory Generation with Diffusion Models for Offline Reinforcement Learning
BiTrajDiff augments offline RL datasets by running independent forward and backward diffusion processes from intermediate states, yielding higher performance than prior one-directional data-augmentation baselines on D4RL.
-
Ada-Diffuser: Latent-Aware Adaptive Diffusion for Decision-Making
Ada-Diffuser is a causal diffusion model that jointly learns observed interaction structure and underlying latent dynamics from minimal observations for adaptive planning and policy learning.
-
TRAP: Tail-aware Ranking Attack for World-Model Planning
TRAP is a tail-aware ranking attack that plants a backdoor in world models so that a trigger causes the model to reorder a few critical imagined trajectories and redirect planning while preserving normal behavior on clean inputs.
-
RAY-TOLD: Ray-Based Latent Dynamics for Dense Dynamic Obstacle Avoidance with TDMPC
RAY-TOLD combines ray-based latent dynamics from LiDAR with MPPI control and a learned policy prior via mixture sampling to lower collision rates in high-density dynamic obstacle environments compared to standard MPPI.
-
Cosmos Policy: Fine-Tuning Video Models for Visuomotor Control and Planning
Single-stage fine-tuning of a video model to generate actions as latent frames plus future states and values yields state-of-the-art robot policy performance on LIBERO, RoboCasa, and bimanual tasks.
-
Ctrl-World: A Controllable Generative World Model for Robot Manipulation
A controllable world model trained on the DROID dataset generates consistent multi-view robot trajectories for over 20 seconds and improves generalist policy success rates by 44.7% via imagined trajectory fine-tuning.
-
Model-Based Reinforcement Learning under Random Observation Delays
A delay-aware model-based RL framework with sequential belief filtering handles random out-of-sequence observations in POMDPs and outperforms MDP baselines while showing robustness to delay shifts.
-
High-Precision and High-Efficiency Trajectory Tracking for Excavators Based on Closed-Loop Dynamics
EfficientTrack integrates model-based learning and closed-loop dynamics to minimize tracking errors in excavator trajectory control with high efficiency and precision, outperforming prior learning-based methods in simulation and real-world tests.
-
V-JEPA 2: Self-Supervised Video Models Enable Understanding, Prediction and Planning
V-JEPA 2 pre-trained on massive unlabeled video achieves strong results on motion understanding and action anticipation, SOTA video QA at 8B scale, and enables zero-shot robotic planning on Franka arms using only 62 hours of unlabeled robot video.
-
Real-Time Execution of Action Chunking Flow Policies
Real-time chunking (RTC) allows diffusion- and flow-based action chunking policies to execute smoothly and asynchronously, maintaining high success rates on dynamic tasks even with significant inference latency.
-
DINO-WM: World Models on Pre-trained Visual Features enable Zero-shot Planning
DINO-WM builds world models on pre-trained DINOv2 features to enable zero-shot planning from offline data without rewards or demonstrations.
-
Valdi: Value Diffusion World Models
Valdi pairs a latent diffusion dynamics model with end-to-end MPC training and reports that one diffusion step matches an MLP baseline on CarRacing while exposing a multimodality-control trade-off.
-
Scaling World-Model Reinforcement Learning Through Diffusion Policy Optimization
MBDPO reformulates policy optimization as a diffusion process over searched trajectories in latent world models to reduce misalignment between search and value learning.
-
stable-worldmodel: A Platform for Reproducible World Modeling Research and Evaluation
The paper presents stable-worldmodel (swm), a platform with high-performance data layer, modern world model baselines, planning solvers, and extended environments for reproducible research and generalization evaluation.
-
D2 Actor Critic: Diffusion Actor Meets Distributional Critic
D2AC combines a diffusion actor with a distributional critic via fused distributional RL and clipped double Q-learning to reach state-of-the-art results on 18 hard control benchmarks including Humanoid, Dog, and Shadow Hand.
-
Self-Predictive Representations for Combinatorial Generalization in Behavioral Cloning
BYOL-γ uses self-predictive representations to approximate successor representations, improving zero-shot combinatorial generalization in goal-conditioned behavioral cloning.
-
Efficient On-policy Visual-RL via Stochastic Decoupled Policy Gradient
SDPG is a new on-policy visual RL algorithm that estimates gradients via stochastic perturbations of rollouts, achieving faster training and lower memory use than baselines on visual MuJoCo tasks while adding new robotics benchmarks and sim-to-real results.
-
Enhancing Human-Likeness in Reinforcement Learning Agents via Hierarchical Macro Action Quantization
HiMAQ applies hierarchical vector quantization to human demonstrations to generate macro actions that yield higher human-likeness scores than flat MAQ on D4RL while matching or exceeding success rates across IQL, SAC, and RLPD.
- Next-Latent Prediction Transformers Learn Compact World Models