Research map / Vision-language and multimodal / Robot action and navigation
Rethinking Progression of Memory State in Robotic Manipulation: An Object-Centric Perspective
Nhat Chung, Taisei Hanyu, Toan Nguyen, Huy Le, Frederick Bumgarner, Duy Minh Ho Nguyen, Khoa Vo, Kashu Yamazaki, Chase Rainwater, Tung Kieu, Anh Nguyen, Ngan Le
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| extend SlotVLA history horizon from 1 to 8 frames | SlotVLA · LIBERO-Goal | success rate (%): 32.0% -> 75.5% | Improved |
| use Naive E-SlotSSM instead of SlotVLA (h=8) | SlotVLA · LIBERO-Goal | success rate (%): 75.5% -> 83.0% | Improved |
| use Naive E-SlotSSM instead of SlotVLA (h=8) on mid drawer open | SlotVLA · LIBERO-Goal | success rate (%): 25% -> 45% | Improved |
| use Naive E-SlotSSM instead of SlotVLA (h=8) on bowl on plate | SlotVLA · LIBERO-Goal | success rate (%): 90% -> 90% | No clear effect |
| use Naive E-SlotSSM instead of pi0 and SlotVLA | baselines · LIBERO-Mem | subgoal completion (%): 5.0% -> 14.8% | Improved |
| extend SlotVLA history horizon from 1 to 8 frames | SlotVLA · LIBERO-Mem | subgoal completion (%): 0% -> 5.0% | Improved |
| use dense-token pi0 (256 tokens) vs slot tokens (SlotVLA h=1) | pi0 · LIBERO-Mem | subgoal completion (%): 0% -> 5.0% | Improved |
| use Naive E-SlotSSM on swap/rotate tasks T7, T8 | SlotSSM · LIBERO-Mem | subgoal completion (%): 0% -> 0% | No clear effect |
| use Naive E-SlotSSM on repeated 7x pick-place T6 and 3x bottle T4 | SlotSSM · LIBERO-Mem | subgoal completion (%): 0% -> 0% | No clear effect |
| use Naive E-SlotSSM with oracle subgoal on occlusion tasks T9 | SlotSSM · LIBERO-Mem | subgoal completion (%): 0% -> 30% | Improved |
Closest papers by meaning
- MemoryVLA: Perceptual-Cognitive Memory in Vision-Language-Action Models for Robotic Manipulation (ICLR 2026 · similarity 0.92)
- RoboMME: Benchmarking and Understanding Memory for Robotic Generalist Policies (ICML 2026 · similarity 0.92)
- EchoVLA: Robotic Vision-Language-Action Model with Synergistic Declarative Memory for Mobile Manipulation (ECCV 2026 · similarity 0.91)
- SAM2Act: Integrating Visual Foundation Model with A Memory Architecture for Robotic Manipulation (ICML 2025 · similarity 0.90)
- SOLD: Slot Object-Centric Latent Dynamics Models for Relational Manipulation Learning from Pixels (ICML 2025 · similarity 0.90)
- ECHO: Continuous Hierarchical Memory for Vision-Language-Action Models (NeurIPS 2026 · similarity 0.89)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
