- Rating
- 2
- Confidence
- 3
- Informativeness
- 2
- Presentation
- 2
- Soundness
- 2
Read review
Paper summary
This paper proposes REKI, a vision-language-action model with short-term memory. It introduces a small set of register tokens in the action expert to store information from past observations. Its attention and normalization design allows these tokens to be reused across frames and denoising steps, supporting batched window training without repeatedly encoding past images. The authors train a 2.2B model on public data and demonstrate improvements on LIBERO-Plus, RoboTwin, and real-robot tasks. Their behavioral analysis also shows that short-term memory reduces back-and-forth motion across action chunks.
Critical strengths
1.The paper studies a meaningful problem. Using historical information while keeping computational costs low is important for continuous robot manipulation.
2.The method is simple and introduces little additional inference cost. Memory is placed directly in the action expert, avoiding repeated VLM processing of past images.
Critical weaknesses
1.Lack of direct comparisons with related memory methods. The paper discusses HAMLET and MemoryVLA but does not directly compare against them. The memoryless baseline demonstrates the value of historical information, but does not establish the advantages of the proposed memory structure over existing methods.
2.Limited ablations of key design choices. The ablations mainly vary the number of historical frames, without separately examining the sampling interval, number of register tokens, gradient propagation through history, or supervision across multiple frames. In particular, although the sampling intervals of 1 second in simulation and 2 seconds on real robots match the action chunk durations, their sensitivity is not evaluated. It therefore remains unclear how robust these settings are across tasks with different motion speeds and periods.
3.Lack of direct evidence for reduced oscillation on real robots. The paper provides oscillation statistics and trajectory analysis in simulation, but the real-robot experiments mainly demonstrate memory’s benefits through task success rates and partial completion scores. These results do not establish whether the gains come from reduced oscillation. Comparisons of oscillation rates and corresponding execution trajectories with and without memory would help verify this effect on real robots.
Questions for authors
Could the authors respond to the concerns raised in the weaknesses section?
4 comments
Answer to ZLGz ctd.
W2. Design ablations
Below we report ablations of the history stride, the number of registers (randomly initialized for a fair comparison) and the supervision of history frames, on RoboCasa365, RoboTwin hard and LIBERO-Plus.
| Ablation | Setting | RoboCasa365 total | RoboTwin hard | LIBERO-Plus |
|---|---|---|---|---|
| Reference | 1 s stride, 32 pretrained registers, all frames supervised | 49.9 | 90.4 | |
| History stride | random in [0.5 s, 2 s] | 51.3 (+1.4) | 90.1 (−0.3) | |
| 2 s | 50.6 (+0.7) | 90.0 (−0.4) | ||
| Supervision | last frame only | 49.3 (−0.6) | 90.8 (+0.4) | |
| Registers per frame (random init.) | 8 | 47.4 (−1.4) | 90.1 (−0.1) | |
| 16 | 47.5 (−1.3) | 90.1 (−0.1) | ||
| 32 | 48.8 | 90.2 | ||
| 64 | 49.4 (+0.6) | 91.0 (+0.8) |
Success rate (%): RoboCasa365 over 50 tasks × 50 episodes, RoboTwin hard over 25 trials per task (35 for the reference), LIBERO-Plus over 10,030 episodes. Small numbers: atomic seen · composite seen · composite unseen. In parentheses: difference with the reference (register rows: with 32 registers).
- Supervising every frame matters. Supervising only the last frame divides the number of supervised action chunks per step by 3, which reduces the effective batch size, but it also makes the task simpler: the model always predicts from the full history, rather than from variable history lengths. This lowers the RoboCasa365 score by 5.0 points (34.2 vs 39.2), while it barely changes the LIBERO-Plus and RoboTwin hard results (+0.4 and −0.6). This suggests that more challenging tasks with larger training datasets, such as RoboCasa365, benefit more from a larger effective batch size, which should matter even more when finetuning on fewer GPUs. We observe the same effect with HAMLET (+3.6 on RoboCasa365 and −1.4 on LIBERO-Plus when every frame is supervised, common answer B).
- History stride. We train with a history stride of 1 s, of 2 s, or sampled at random in [0.5 s, 2 s] for each window. At inference, all three models replan every 1 s, and the model trained with random strides uses a stride of 1 s. The effect of the stride varies across benchmarks, but on average all three models perform similarly (−0.3 points with random strides and +0.5 with a 2 s stride, against the reference), which shows that the memory mechanism is robust to the choice of history stride. The stride can therefore be chosen depending on the pace of the task.
- Registers. With randomly initialized registers, performance increases from 8 to 64 registers from 37.4 to 41.2 on RoboCasa365, from 47.4 to 49.4 on RoboTwin hard and from 90.1 to 91.0 on LIBERO-Plus, suggesting that the memory benefits from more capacity per frame.
Note that evaluating the same checkpoint several times gives results that vary by about 1 point (common answer B), from numerical non-determinism in compiled inference. We will add this table and its discussion to our revision.
W3. Oscillation on the real robot
We applied the oscillation detector, see our reply to 5tf5 to the recorded real-robot runs, on the 5 tasks of Fig. 7 (success counts in our reply to qRzx), with the same thresholds as in simulation.
| Task | Oscillation (%) ↓ Without | Oscillation (%) ↓ With | Task score (%) ↑ Without | Task score (%) ↑ With |
|---|---|---|---|---|
| Stack three cubes | 40.0 (8/20) | 13.6 (3/22) | 35 | 50 |
| Sweep paper into dustpan | 27.3 (3/11) | 5.9 (1/17) | 57.5 | 57.5 |
| Place 3 fruits in plates | 0.0 (0/10) | 18.2 (2/11) | 48 | 60 |
| Place banana in box | 0.0 (0/8) | 0.0 (0/28) | 65 | 65 |
| Stack three cups | 9.1 (1/11) | 4.8 (1/21) | 67.5 | 85 |
| Average | 15.3 | 8.5 | 54.6 | 63.5 |
Without / with memory. Oscillation: share of episodes with oscillation (oscillating / recorded episodes). Task score: Fig. 7.
Memory reduces the average oscillation rate by 1.8×, from 15.3% to 8.5%. Most of the reduction comes from stacking cubes and sweeping, the two tasks where the model without memory oscillates most.
We will add these rates and the joint trajectories with and without memory to our revised version.
Thank you for the detailed responses and the time spent conducting these additional experiments. The comparisons with existing memory methods, expanded ablations, and real-robot oscillation analysis have substantially addressed my main concerns. I appreciate these efforts and will take the additional results and clarifications into consideration when determining my final score.
Response to Reviewer ZLGz
We thank the reviewer for their constructive feedback, which helped us strengthen the paper, and for taking the additional results into consideration.
Answer to ZLGz
W1. Comparison with HAMLET and MemoryVLA
We ported HAMLET and MemoryVLA from the authors' released code to REKI. We now compare REKI with HAMLET (as published and with the all-frame supervision of REKI), MemoryVLA, a feature cache and frame stacking, on the same backbone and in the same training pipeline; for details, please see the common answer above (1/3, 2/3, 3/3). In short: