mlsanityData to 7 Oct 2026

Memory, Benchmark & Robots: A Benchmark for Solving Complex Tasks with Reinforcement Learning

arXiv 2502.10550 ↗First posted 14 Feb 2025cs.LG cs.AI cs.ROcited by 34

Not on the map. Why?

Abstract (arXiv)

Memory is crucial for enabling agents to tackle complex tasks with temporal and spatial dependencies. While many reinforcement learning (RL) algorithms incorporate memory, the field lacks a universal benchmark to assess an agent's memory capabilities across diverse scenarios. This gap is particularly evident in tabletop robotic manipulation, where memory is essential for solving tasks with partial observability and ensuring robust performance, yet no standardized benchmarks exist. To address this, we introduce MIKASA (Memory-Intensive Skills Assessment Suite for Agents), a comprehensive benchmark for memory RL, with three key contributions: (1) we propose a comprehensive classification framework for memory-intensive RL tasks, (2) we collect MIKASA-Base -- a unified benchmark that enables systematic evaluation of memory-enhanced agents across diverse scenarios, and (3) we develop MIKASA-Robo (pip install mikasa-robo-suite) -- a novel benchmark of 32 carefully designed memory-intensive tasks that assess memory capabilities in tabletop robotic manipulation. Our work introduces a unified framework to advance memory RL research, enabling more robust systems for real-world use. MIKASA is available at https://tinyurl.com/membenchrobots.

Key facts. Cited by 34 papers in the map (data through 7 Oct 2026). Builds on 71 of its 96 references that are in the map. First posted on arXiv 14 Feb 2025.

Builds on

71 of 96 references.

References · 71

  1. 2017Attention Is All You Needcited by 59,932
  2. 2017Proximal Policy Optimization Algorithmscited by 14,441
  3. 2018Soft Actor-Critic: Off-Policy Maximum Entropy Deep Reinforcement Learning with a Stochastic Actorcited by 4,635
  4. 2023Mamba: Linear-Time Sequence Modeling with Selective State Spacescited by 4,616
  5. 2023Diffusion Policy: Visuomotor Policy Learning via Action Diffusioncited by 2,730
  6. 2021Efficiently Modeling Long Sequences with Structured State Spacescited by 2,424
  7. 2014Empirical Evaluation of Gated Recurrent Neural Networks on Sequence Modelingcited by 2,341
  8. 2020Conservative Q-Learning for Offline Reinforcement Learningcited by 1,777
  9. 2018Graph Neural Networks: A Review of Methods and Applicationscited by 1,571
  10. 2021Decision Transformer: Reinforcement Learning via Sequence Modelingcited by 1,528
  11. 2023LongBench: A Bilingual, Multitask Benchmark for Long Context Understandingcited by 1,378
  12. 2018Learning Latent Dynamics for Planning from Pixelscited by 1,221
38 more
  1. 2019Meta-World: A Benchmark and Evaluation for Multi-Task and Meta Reinforcement Learningcited by 1,168
  2. 2021Isaac Gym: High Performance GPU-Based Physics Simulation For Robot Learningcited by 882
  3. 2017AI2-THOR: An Interactive 3D Environment for Visual AIcited by 850
  4. 2022Simplified State Space Layers for Sequence Modelingcited by 776
  5. 2019RLBench: The Robot Learning Benchmark & Learning Environmentcited by 669
  6. 2021CLIPort: What and Where Pathways for Robotic Manipulationcited by 617
  7. 2021Habitat 2.0: Training Home Assistants to Rearrange their Habitatcited by 501
  8. 2020robosuite: A Modular Simulation Framework and Benchmark for Robot Learningcited by 484
  9. 2023TD-MPC2: Scalable, Robust World Models for Continuous Controlcited by 482
  10. 2019Introduction to Multi-Armed Banditscited by 475
  11. 2015Deep Recurrent Q-Learning for Partially Observable MDPscited by 457
  12. 2022VIMA: General Robot Manipulation with Multimodal Promptscited by 403
  13. 2024RoboCasa: Large-Scale Simulation of Everyday Tasks for Generalist Robotscited by 391
  14. 2019Relay Policy Learning: Solving Long-Horizon Tasks via Imitation and Reinforcement Learningcited by 380
  15. 2020dm_control: Software and Tasks for Continuous Controlcited by 368
  16. 2024Prismatic VLMs: Investigating the Design Space of Visually-Conditioned Language Modelscited by 336
  17. 2016Learning Hand-Eye Coordination for Robotic Grasping with Deep Learning and Large-Scale Data Collectioncited by 312
  18. 2019Stabilizing Transformers for Reinforcement Learningcited by 262
  19. 2018BabyAI: A Platform to Study the Sample Efficiency of Grounded Language Learningcited by 238
  20. 2023L-Eval: Instituting Standardized Evaluation for Long Context Language Modelscited by 232
  21. 2024ManiSkill3: GPU Parallelized Robotics Simulation and Rendering for Generalizable Embodied AIcited by 216
  22. 2020The NetHack Learning Environmentcited by 196
  23. 2021iGibson 2.0: Object-Centric Simulation for Robot Learning of Everyday Household Taskscited by 191
  24. 2024BEHAVIOR-1K: A Human-Centered, Embodied AI Benchmark with 1,000 Everyday Activities and Realistic Simulationcited by 140
  25. 2019Behaviour Suite for Reinforcement Learningcited by 129
  26. 2021Recurrent Model-Free RL Can Be a Strong Baseline for Many POMDPscited by 115
  27. 2020CausalWorld: A Robotic Manipulation Benchmark for Causal Structure and Transfer Learningcited by 109
  28. 2023Structured State Space Models for In-Context Reinforcement Learningcited by 109
  29. 2021MiniHack the Planet: A Sandbox for Open-Ended Reinforcement Learning Researchcited by 96
  30. 2017Neural Map: Structured Memory for Deep Reinforcement Learningcited by 88
  31. 2022Partially Observable Markov Decision Processes in Robotics: A Surveycited by 81
  32. 2019Meta reinforcement learning as task inferencecited by 75
  33. 2023ARNOLD: A Benchmark for Language-Grounded Task Learning With Continuous States in Realistic 3D Scenescited by 68
  34. 2023When Do Transformers Shine in RL? Decoupling Memory from Credit Assignmentcited by 62
  35. 2016Control of Memory, Active Perception, and Action in Minecraftcited by 61
  36. 2023POPGym: Benchmarking Partially Observable Reinforcement Learningcited by 59
  37. 2025SAM2Act: Integrating Visual Foundation Model with A Memory Architecture for Robotic Manipulationcited by 58
  38. 2018Optimizing Agent Behavior over Long Time Scales by Transporting Valuecited by 57

50 most cited of 71 All 71 (JSON)

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2025: 7; 2026: 27.

34 papers.

Sort:

By area:

  1. 2025MemoryVLA: Perceptual-Cognitive Memory in Vision-Language-Action Models for Robotic Manipulationcited by 183
  2. 2025A Survey of Context Engineering for Large Language Modelscited by 120
  3. 2025Large VLM-based Vision-Language-Action Models for Robotic Manipulation: A Surveycited by 66
  4. 2026RMBench: Memory-Dependent Robotic Manipulation Benchmark with Insights into Policy Designcited by 40
  5. 2025Don't Blind Your VLA: Aligning Visual Representations for OOD Generalizationcited by 34
  6. 2025Rethinking Progression of Memory State in Robotic Manipulation: An Object-Centric Perspectivecited by 20
  7. 2026EventVLA: Event-Driven Visual Evidence Memory for Long-Horizon Vision-Language-Action Policiescited by 18
  8. 2026Gated Memory Policy: In-Context Memorization and Adaptationcited by 13
  9. 2026MemoryVLA++: Temporal Modeling via Memory and Imagination in Vision-Language-Action Modelscited by 9
  10. 2026vla-eval: A Unified Evaluation Harness for Vision-Language-Action Modelscited by 7
  11. 2026μVLA: On Recurrent Memory for Partially Observable Manipulation in VLA Modelscited by 7
  12. 2025ELMUR: External Layer Memory with Update/Rewrite for Long-Horizon RL Problemscited by 6
22 more
  1. 2025NEBULA: Do We Evaluate Vision-Language-Action Agents Correctly?cited by 6
  2. 2026VPWEM: Non-Markovian Visuomotor Policy with Working and Episodic Memorycited by 6
  3. 2026Chameleon: Control-Indexed Prospective Memory for Visuomotor Manipulationcited by 4
  4. 2026Data Pyramid for Embodied Manipulation: A Surveycited by 4
  5. 2026SimpleMemVLA: A Simple but Effective Native-Video Memory for Vision-Language-Action Modelscited by 2
  6. 2026Memory Retention Is Not Enough to Master Memory Tasks in Reinforcement Learningcited by 1
  7. 2026WatchAct: A Benchmark for Behavior-Grounded Robot Manipulationcited by 1
  8. 2026FibVLA: An Efficient Temporal Vision-Language-Action Model with Fibonacci Samplingcited by 1
  9. 2026KAGE-Bench: Fast Known-Axis Visual Generalization Evaluation for Reinforcement Learningcited by 0
  10. 2026Does VLA Even Know the Basics? Measuring Commonsense and World Knowledge Retention in Vision-Language-Action Modelscited by 0
  11. 2026RoboSPA: Can VLA Models Go Beyond Simple Scenes and Short-Horizon Tasks?cited by 0
  12. 2026Learning to Use Imagination: Progress-Conditioned Future Utilization for World Action Modelscited by 0
  13. 2026MEMOBench: A Process Level Memory Benchmark for Robotic Manipulationcited by 0
  14. 2026SmoLSTM: A Compact Vision-Language-Action Model with Recurrent Memory that Persistscited by 0
  15. 2026ActiveArena: Benchmarking and Understanding Active Perception in Robotic Manipulationcited by 0
  16. 2026Minimal Recurrent Behavioral Memory for Imitation under Partial Observabilitycited by 0
  17. 2026MemBodied: Recurrent Associative Memory for Vision-Language-Action Modelscited by 0
  18. 2026RoboIRGBench: Benchmarking Implicit Referential Grounding in Vision-Language-Action Modelscited by 0
  19. 2026Simple Agentic Memory for Generalist Robot Policiescited by 0
  20. 2026ALER: Adaptive Learnable Experience Rewriting for Reinforcement Learningcited by 0
  21. 2026Decision Titan: Test-Time Training for Long-Term Memory in Offline Reinforcement Learningcited by 0
  22. 2026Self-Supervised Keyframe Discovery for Horizon-Invariant Behavior Cloningcited by 0