- Rating
- 2
- Confidence
- 2
- Informativeness
- 2
- Presentation
- 3
- Soundness
- 2
Read review
Paper summary
This paper addresses the phenomenon that failure in embodied tasks may stem from inherent skill deficiencies or the execution lapse to follow existing effective guidance. It proposes EmbodiSkill, a method that requires no training and no updating of model parameters. Its skill-aware reflection categorizes trajectory evidence into Discovery, Optimization, Skill Defect, and Execution Lapse. The first three categories are used to update the skill text; the Execution Lapse is used to update the independent appendix, emphasizing rules already present but ignored in the text.
Experimental results on ALFWorld, EB-Habitat, and EB-Navigation demonstrate that EmbodiSkill effectively enhances the embodiment capabilities of general multimodal models. The authors also provide analysis of reflection and appendix ablation, evolutionary processes, hyperparameter sensitivity analysis, and cue templates.
Critical strengths
- The problem is practically relevant, and the method design corresponds to the problem. It distinguishes between knowledge that needs modification and re-emphasis, making it suitable for long-term skill library maintenance.
- Fully local configuration also offers significant gains. Multiple models on ALFWorld and EB-Habitat show substantial improvements, indicating that the method does not entirely rely on closed-source evolvers.
Critical weaknesses
- The core novelty has not yet been fully established. While the AutoManual already addresses failure attribution and rule emphasis[1], this paper needs to more directly demonstrate the unique advantages of independently maintaining the appendix.
- Ablation and validation remain insufficient. Existing experiments cannot independently demonstrate the role of failure attribution and the independent appendix, and validation of attribution accuracy, repeatability, and statistical stability is lacking.
References:
[1]Automanual: Constructing instruction manuals by llm agents via interactive environmental learning. In NeurIPS 2024.
Questions for authors
- AutoManual already includes failure attribution and rule emphasis. What has been added in this article? Can we compare in-rule emphasis and independent appendixes under the same executor and budget?
- "w/o Skill-Aware Reflection" removes both reflection and failure attribution. Can we retain reflection and only remove failure categories to verify the effectiveness of attribution itself?
2 comments
Q1. Compared to AutoManual, what are the additional contributions of EmbodiSkill? (part 2)
New direct comparison experiments further demonstrate the overall advantages of EmbodiSkill over AutoManual. As shown in Tables 1 and 2, under the same executors and evolution budgets, EmbodiSkill consistently outperforms AutoManual across ALFWorld, EB-Habitat, and EB-Navigation, with improvements of up to 16.42, 11.00, and 10.00 percentage points, respectively.
Table 1&2 in the paper. Direct comparison between AutoManual and EmbodiSkill across embodied benchmarks. All values are task success rates (%).
| Benchmark | Executor | AutoManual | EmbodiSkill (Ours) |
|---|---|---|---|
| ALFWorld | Qwen2.5-14B | 63.43 | 79.85 |
| ALFWorld | Qwen3.5-27B | 73.13 | 86.57 |
| EB-Habitat | Qwen3-VL-8B | 31.67 | 42.67 |
| EB-Habitat | Qwen3-VL-32B | 48.00 | 51.33 |
| EB-Navigation | Qwen3-VL-8B | 39.00 | 49.00 |
| EB-Navigation | Qwen3-VL-32B | 49.67 | 56.00 |
Q2. How Can We Independently Validate the Effectiveness of Failure Attribution?
Highlight. We provide two key findings regarding Failure Attribution: (1) Effectiveness: Removing Failure Attribution while retaining Reflection reduces success from 85.82% to 54.48%, even below removing Reflection entirely (67.91%), demonstrating that incorrect attribution can be more harmful than omitting Reflection. (2) Reliability: Human evaluation achieves 96% agreement, supporting the reliability of distinguishing Skill Defects from Execution Lapses.
We introduce a new w/o Failure Attribution ablation to independently evaluate the contribution of Failure Attribution while retaining Reflection. This variant preserves Discovery and Optimization but treats all failed-trajectory reflections as Skill Defects, using them to revise the Skill Body without updating the Appendix. As shown in Table 3, with the same Gemini-3-flash evolution model and Qwen2.5-14B executor, removing Failure Attribution reduces success from 85.82% to 54.48% (−31.34 percentage points), even below w/o Skill-Aware Reflection (67.91%). This suggests that incorrect attribution can be more harmful than omitting Reflection, as execution lapses may trigger unnecessary revisions to correct guidance. The GPT results show a similar numerical trend (76.12% vs. 86.57%), although the evolution models differ.
Table 3 in the paper. Ablation study of Failure Attribution on ALFWorld.
| Configuration | GPT | Gemini |
|---|---|---|
| EmbodiSkill | 86.57 | 85.82 |
| -w/o Skill-Aware Reflection | 70.90 | 67.91 |
| -w/o Failure Attribution | 76.12 | 54.48 |
| -w/o Appendix | 76.87 | 78.36 |
Note: As GPT-5.2 is no longer accessible, the new w/o Failure Attribution experiment uses GPT-5.6-sol, while the other experiments in the GPT column use GPT-5.2.
We further conduct an independent human evaluation to assess the reliability of Failure Attribution. We randomly sampled 50 failed-trajectory reflections and independently annotated them using complete trajectories and existing Skills, assessing whether following the original guidance could have prevented failure. As shown in Table 5, the model agrees with human annotations on 48/50 cases (96%). These results demonstrate the reliability of the model's failure attribution.
Table 5 in the paper. Comparison between human annotations and model predictions for Failure Attribution.
| Human Label | Pred. Defect | Pred. Lapse | Total |
|---|---|---|---|
| Skill Defect | 14 | 0 | 14 |
| Execution Lapse | 2 | 34 | 36 |
| Total | 16 | 34 | 50 |
We sincerely thank the reviewer again for their valuable feedback. We hope that the additional experiments and clarifications adequately address the reviewer's concerns and further strengthen the paper.
References
[1] Chen et al. (2024). AutoManual: Constructing Instruction Manuals by LLM Agents via Interactive Environmental Learning. NeurIPS 2024. https://arxiv.org/abs/2405.16247
[2] Liu et al. (2024). Lost in the Middle: How Language Models Use Long Contexts. TACL, 12:157–173. https://aclanthology.org/2024.tacl-1.9/
[3] Leviathan et al. (2025). Prompt Repetition Improves Non-Reasoning LLMs. arXiv:2512.14982. https://arxiv.org/abs/2512.14982
We sincerely thank the reviewer for their constructive feedback. We appreciate the reviewer's positive assessment of the practical relevance of our problem, the suitability of our approach for long-term skill maintenance, and the substantial performance improvements demonstrated in our experiments. The reviewer's important concerns regarding the independent Appendix and Failure Attribution have helped us strengthen the paper. We have incorporated additional experiments and analyses into the revised manuscript, with all changes highlighted in blue. We address these concerns in detail below.
Q1. Compared to AutoManual, what are the additional contributions of EmbodiSkill? (part 1)
EmbodiSkill enables continuous self-evolution through multi-trajectory experience. In embodied tasks, the same goal may require different action sequences across environments, making updates based on individual trajectories prone to overfitting to the specific environment in which they were collected. Unlike AutoManual[1], which updates rules after each individual trajectory, EmbodiSkill employs Skill-Aware Reflection and consolidates experience across multiple trajectories before updating skills. The updated skills are then reused in subsequent interactions, forming a continuous cycle of reflection, consolidation, and skill evolution.
Our newly added Table 7 provides supporting evidence. Under the same 500-trajectory evolution budget (K=1), batch updates with B=50 and B=100 achieve 79.85% and 73.88% success, respectively, outperforming per-trajectory updates (B=1, 61.94%) by 17.91 and 11.94 percentage points.
Table 7 in the paper. Sensitivity to the number of trajectories per update B and reflection limit K. The total evolution budget is fixed at 500 trajectories, with N = 500/B evolution stages.
The independent Appendix is a targeted reinforcement mechanism designed for local embodied executors. Given the latency requirements of embodied interaction, frozen local models are commonly employed as executors. Prior studies have shown that LLMs exhibit positional biases, often underutilizing information in the middle of a prompt [2], while prompt repetition can improve model performance in certain settings [3]. Consequently, AutoManual's in-place bolding or annotations may still leave critical guidance vulnerable to positional biases or insufficient repetition. Instead, EmbodiSkill introduces an independent Appendix following the skill body, where overlooked guidance identified by Skill-Aware Reflection is selectively reiterated. The Appendix therefore enhances its salience and reinforces its influence during execution, mitigating the mentioned limitation.
Our newly added Table 4 directly validates this design. With the same Skill, emphasized content, and executor, our Appendix achieves 93.28% success, outperforming Inline Bolding and Inline Reminder by 6.71 and 8.95 percentage points, respectively. It also surpasses Random Reminders (79.85%), Full-Body Repetition (82.84%), Top-k Body Retrieval (85.07%), and Top-k Appendix Retrieval (83.58%), demonstrating the effectiveness of our proposed design.
Table 4 in the paper. Comparison of skill reinforcement strategies on ALFWorld.