- Rating
- 3
- Confidence
- 2
- Informativeness
- 3
- Presentation
- 2
- Soundness
- 3
Read review
Paper summary
This paper introduces a lifelong learning approach that enables embodied intelligence systems to continuously accumulate experience and skills. It accumulates and compresses knowledge in symbolic form to iteratively improve the overall system’s performance. The method’s effectiveness is validated on multiple benchmarks and through a limited set of real-world robot experiments.
Critical strengths
- Summarizing, accumulating, and compressing experience using textual symbols is an effective and reasonable approach that makes full use of language models’ strengths.
- The experimental validation is generally thorough and supports the paper’s claims.
- The paper provides extensive visualizations and supporting details, along with real-world robot experiments that demonstrate the method’s effectiveness.
Critical weaknesses
- The generalizability and reusability of the accumulated knowledge need to be clarified, particularly which knowledge can be shared across different tasks or even different benchmarks.
- The efficiency of knowledge accumulation and the additional overhead of invoking the corresponding harness during execution may warrant dedicated analysis.
- If the knowledge itself takes the form of a harness, the paper does not clarify whether its benefits generalize across different underlying execution models.
Questions for authors
- Provide a detailed analysis of whether the accumulated knowledge generalizes across different tasks, benchmarks, and underlying execution models.
- Conduct a dedicated analysis of the efficiency of knowledge accumulation and the overhead of invoking the corresponding harness, and draw clear conclusions from the results.
4 comments
W2 Q2. The efficiency of knowledge accumulation and the additional overhead of invoking the corresponding harness during execution warrant a dedicated analysis with clear conclusions.
To address this concern, we first examine how symbolic knowledge and compiled neural memory accumulate, together with the time and memory costs of the two learning processes. We then analyze how the shift from symbolic reasoning to compiled neural memory affects inference efficiency. The table below reports accumulated experiences, retained knowledge sizes, memory usage, and wall-clock time for Knowledge Consolidation and Knowledge Compilation.
| Round | Accumulated experiences (# traces / # transitions) | │ | Knowledge Consolidation | │ | Knowledge Compilation | ||||
|---|---|---|---|---|---|---|---|---|---|
| │ | Symbolic knowledge size (#) | Host memory (GB) | Wall-clock time (h) | │ | Adapter-pool size (#) | GPU memory (GB) | Wall-clock time (h) | ||
| 1 | 20 / 96 | │ | 668 | 11.04 | 0.65 | │ | 156 | 18.15 | 1.55 |
| 2 | 40 / 180 | │ | 674 | 11.39 | 0.13 | │ | - | - | - |
| 3 | 60 / 213 | │ | 755 | 3.28 | 0.31 | │ | 364 | 15.25 | 2.26 |
| 4 | 80 / 337 | │ | 763 | 8.75 | 4.19 | │ | - | - | - |
| 5 | 100 / 475 | │ | 767 | 10.84 | 0.85 | │ | 546 | 15.54 | 5.73 |
| 6 | 120 / 633 | │ | 767 | 7.26 | 0.21 | │ | - | - | - |
| 7 | 140 / 757 | │ | 774 | 9.19 | 0.56 | │ | 728 | 17.65 | 1.98 |
| 8 | 160 / 887 | │ | 778 | 8.45 | 1.16 | │ | - | - | - |
| 9 | 180 / 992 | │ | 787 | 12.53 | 1.82 | │ | 1,014 | 17.66 | 3.46 |
| 10 | 200 / 1,094 | │ | 776 | 9.53 | 0.69 | │ | 1,300 | 17.65 | 3.38 |
Symbolic knowledge size counts the constituent symbolic elements, including predicates, parameters, and action names, while adapter-pool size counts individual semantic and procedural adapter modules. Host and GPU memory report peak usage during consolidation and adapter training, respectively.
Symbolic knowledge size changes little after the early rounds, while neural memory continues to expand as more reasoning patterns are compiled. The reported wall-clock times total 10.57 h for ten consolidation rounds and 18.36 h for six compilation rounds. Although cumulative learning costs increase, per-round learning time and peak memory usage show no consistent upward trend within the evaluated stream. Both learning processes operate in the background, separately from online inference.
We also summarize the relative use of consolidated symbolic knowledge and compiled neural memory across learning rounds, together with inference time on seen and unseen tasks, using the results reported in Figure 3 of the main paper. This analysis examines how the increasing use of compiled neural memory relates to inference efficiency as knowledge accumulates.
| Round | Consolidated symbolic knowledge inference (%) | Compiled neural memory inference (%) | Inference time, seen (s) | Inference time, unseen (s) |
|---|---|---|---|---|
| 0 | 100.00 | 0.00 | 7.64 | 4.77 |
| 1 | 92.74 | 7.26 | 6.02 | 6.22 |
| 2 | 80.28 | 19.72 | 5.22 | 4.43 |
| 3 | 69.74 | 30.26 | 3.35 | 4.09 |
| 4 | 59.26 | 40.74 | 3.33 | 3.76 |
| 5 | 45.54 | 54.46 | 3.13 | 2.97 |
| 6 | 30.23 | 69.77 | 2.95 | 2.78 |
| 7 | 17.00 | 83.00 | 0.92 | 0.79 |
| 8 | 8.04 | 91.96 | 0.75 | 0.57 |
| 9 | 0.55 | 99.45 | 0.61 | 0.31 |
| 10 | 0.00 | 100.00 | 0.21 | 0.31 |
Under Default inference, the share of inference handled through symbolic reasoning decreases from 100.00% at Round 0 to 0.00% at Round 10, with compiled neural memory becoming the dominant inference path from Round 5 onward. During this transition, inference time decreases from 7.64 s to 0.21 s on seen tasks and from 4.77 s to 0.31 s on unseen tasks, falling below 1 s on both task sets from Round 7 onward. Over the same rounds, task success improves from 26.14% to 81.05% on seen tasks and from 27.19% to 73.25% on unseen tasks.
Together, these analyses show that knowledge accumulation incurs background learning costs, while Knowledge Compilation reduces inference time by enabling the reuse of validated reasoning patterns with less reliance on symbolic planning.
In response to the reviewer's thoughtful feedback, we have added experimental analyses to the Appendix. Appendix A.5.7 presents the additional evaluation of accumulated knowledge reuse across different underlying execution models. Appendix A.5.5 reports knowledge accumulation and the associated time and memory costs of Knowledge Consolidation and Knowledge Compilation. Toward the end of the discussion phase, once our discussion has concluded, we will incorporate the relevant findings and clarifications into the main paper.
Reply to the authors
Thanks for your reply. The new details address some of my concerns. But due to the limitation of direct related experience, I still keep my rating with low confidence.
Thank you for your response and for considering our clarifications. We would be happy to address any further questions or concerns. We sincerely appreciate your time and valuable feedback.
We greatly thank the reviewer for the constructive feedback. We address each point below and indicate the corresponding paper revisions. We would be happy to clarify any remaining concerns.
The real-world experiments in Figure 4 provide concrete examples of how accumulated symbolic knowledge generalizes and is reused across tasks and from CaP-Bench simulation to real-world deployment.
Figure 4(a) shows how symbolic knowledge acquired in CaP-Bench is extended to handle additional operating constraints in real-world task settings. The agent reuses its existing knowledge for object placement and appliance manipulation while refining the conditions for appliance activation using reference manuals. For the induction stove, it adds
stove_compatible(obj)andon_top_of(obj, stove)as preconditions, requiring compatible cookware to be placed on the cooking zone before activation. For the air fryer, it addstemperature-set(off)andtimer-set(off)as preconditions, requiring both knobs to be reset before activation. These updates extend the transferred knowledge to operating conditions not covered by the original simulation experience, while retaining existing knowledge for the remaining task steps. Appendix B.2.1 provides further details.Figure 4(b) further demonstrates the compositional generalizability of newly acquired symbolic knowledge and the reusability of existing symbolic knowledge. Across successive task executions, the agent reuses knowledge supporting previously successful behaviors and combines it with new knowledge from different sources to develop new task-solving strategies and progressively improve task quality. Feedback from its own execution first refines
Pushby addingmoved(obj, loc)to its effects, enabling keyboard relocation. TheStackschema acquired from another robot's cube-stacking task demonstration is then reused to stack plastic cups in the desk-cleaning task, while the previously acquired pushing behavior remains in use. Human-provided knowledge subsequently introducesBin_presswithfree_space(bin)as an effect, enabling additional objects to be discarded into a previously full bin while retaining both pushing and stacking.These schemas are used together with the existing
Pick,Place, andWipeschemas, allowing the agent to relocate objects, stack items compactly, and free up bin space for further disposal. The resulting strategies expose and clean progressively larger areas of the desk mat, improving task quality beyond successful completion. Appendix B.2.2 and Figure 19 illustrate these executions.Reuse across underlying execution models is addressed in our response to W3 below.
We agree that the consolidated symbolic knowledge serves as a harness for the underlying execution model. Its predicates and action schemas are represented independently of model parameters, allowing the same symbolic knowledge to support symbolic reasoning across different execution models.
To evaluate whether these benefits generalize across models, we use the same accumulated symbolic knowledge, tasks, and symbolic tools for all execution models at each evaluated checkpoint. We replace the underlying execution model and adjust the Model Interface implementation for prompt formatting and output parsing accordingly. Since neural adapters are model-specific, we recompile the validated reasoning patterns into neural memory for each execution model, using the attachment layers and representation dimensions configured for that model.
Replacing the underlying execution model preserves task performance close to the reference configuration while maintaining reduced symbolic tool reliance after Block 5 and 10. Across the evaluated models, final success rates range from 90.50% to 91.33%, while symbolic tool reliance decreases from an initial 100.00% to between 67.23% and 72.13%. These results indicate that the benefits of accumulated symbolic knowledge and its model-specific compilation extend across the evaluated execution models. The symbolic predicates and action schemas provide reusable support across models, while Knowledge Compilation enables each model to generate outputs for validated reasoning patterns through its own adapters, reducing reliance on external symbolic tools.