Reviewer #1
Questions
2. I am an expert on the topic of the paper.
Agree
3. The title and abstract reflect the content of the paper.
Strongly agree
4. The paper discusses, cites and compares with all relevant related work
Agree
5. Please justify the previous choice (Required if “Strongly Disagree” or “Disagree” is chosen, otherwise write "n/a")
n/a
6. Readability and paper organization: The writing and language are clear and structured in a logical manner.
Strongly agree
7. The paper adheres to ISMIR 2026 submission guidelines (uses the ISMIR 2026 template, has at most 6 pages of technical content followed by “n” pages of references, AI usage declaration or ethical considerations, references are well formatted). If you selected “No”, please explain the issue in your comments.
Yes
8. Relevance of the topic to ISMIR: The topic of the paper is relevant to the ISMIR community. Note that submissions of novel music-related topics, tasks, and applications are highly encouraged. If you think that the paper has merit but does not exactly match the topics of ISMIR, please do not simply reject the paper but instead communicate this to the Program Committee Chairs. Please do not penalize the paper when the proposed method can also be applied to non-music domains if it is shown to be useful in music domains.
Strongly agree
9. Scholarly/scientific quality: The content is scientifically correct.
Agree
10. Please justify the previous choice (Required if "Strongly Disagree" or "Disagree" is chosen, otherwise write "n/a")
n/a
11. Novelty of the paper: The paper provides novel methods, applications, findings or results. Please do not narrowly view "novelty" as only new methods or theories. Papers proposing novel musical applications of existing methods from other research fields are considered novel at ISMIR conferences.
Agree
12. The paper provides all the necessary details or material to reproduce the results described in the paper. Keep in mind that ISMIR respects the diversity of academic disciplines, backgrounds, and approaches. Although ISMIR has a tradition of publishing open datasets and open-source projects to enhance the scientific reproducibility, ISMIR accepts submissions using proprietary datasets and implementations that are not sharable. Please do not simply reject the paper when proprietary datasets or implementations are used.
Agree
13. Pioneering proposals: This paper proposes a novel topic, task or application. Since this is intended to encourage brave new ideas and challenges, papers rated "Strongly Agree" and "Agree" can be highlighted, but please do not penalize papers rated "Disagree" or "Strongly Disagree". Keep in mind that it is often difficult to provide baseline comparisons for novel topics, tasks, or applications. If you think that the novelty is high but the evaluation is weak, please do not simply reject the paper but carefully assess the value of the paper for the community.
Agree (Novel topic, task, or application)
14. Reusable insights: The paper provides reusable insights (i.e. the capacity to gain an accurate and deep understanding). Such insights may go beyond the scope of the paper, domain or application, in order to build up consistent knowledge across the MIR community.
Strongly agree
15. Please explain your assessment of reusable insights in the paper.
With diffusion-type model being applied to music editing, this is an important topic to explore. While some specific scores presented in this evaluation scheme may not perfectly correlate with subjective judgment, the overall framework and division of categories in the multi-facet framework can inspire follow-up works.
16. AI Usage Policy: The paper complies with the ISMIR 2026 AI Usage Policy for Authors. (Authors are welcome to use any tool they wish for preparing and writing the paper but they must ensure that all content is correct and original. Authors must declare any use of LLMs and AI tools if it's a part of core methodology (including literature review, related work, generated figures, tables or other illustrations). Using AI tools for editing purposes (e.g,. checking grammar and fixing typos) need not be declared. Please pay attention to any hallucinated text, figures, or references and flag them in the comments below)
Agree
17. Please justify the previous choice (Required if "Disagree" is chosen, otherwise write "n/a")
n/a
18. Write ONE line (in your own words) with the main take-home message from the paper.
This paper contributes a set of metrics to evaluate whether context (i.e. content that was not supposed to be modified) would be preserved in music editing. They evaluated it using some objective editing (where they have ground truth) and some (diffusion-based) editing models, where they do not have ground truth but is closer to real world applications.
21. Potential to generate discourse: The paper will generate discourse at the ISMIR conference or have a large influence/impact on the future of the ISMIR community.
Strongly agree
22. Overall evaluation: Keep in mind that minor flaws can be corrected, and should not be a reason to reject a paper. Please familiarize yourself with the reviewer guidelines at https://ismir.net/reviewer-guidelines
Weak accept
23. Main review and comments for the authors. Please summarize strengths and weaknesses of the paper. It is essential that you justify the reason for the overall evaluation score in detail. Keep in mind that belittling or sarcastic comments are not appropriate.
The paper identifies an interesting research question and is tightly connected to mainstream popular music generation models.

I noticed a few potential issues:
- the Melodic content/motif category behaved a bit unpredictably in both objective (Table 2) and subjective (Table 3) evaluations. Looks like changing semitones shift melodic motif greatly, which sounds a bit flawed to me. A few more metrics seem to incompletely capture human perception, which makes sense because the scores feel like summary statistics and may not fully capture the complexity of human perception.

- While the objective evaluation (Section 3) mostly focused on controlled semitone/rhythm/melody edits, the application on music edit models involve style transfer and tasks that were not evaluated by section 3, even though I found it technically not difficult to also evaluate style/instrument transfer by manual editing. This makes the interpretation of the 4-model case studies quite limited in my opinion.
Reviewer #2
Questions
2. I am an expert on the topic of the paper.
Strongly agree
3. The title and abstract reflect the content of the paper.
Strongly agree
4. The paper discusses, cites and compares with all relevant related work
Agree
5. Please justify the previous choice (Required if “Strongly Disagree” or “Disagree” is chosen, otherwise write "n/a")
n/a
6. Readability and paper organization: The writing and language are clear and structured in a logical manner.
Agree
7. The paper adheres to ISMIR 2026 submission guidelines (uses the ISMIR 2026 template, has at most 6 pages of technical content followed by “n” pages of references, AI usage declaration or ethical considerations, references are well formatted). If you selected “No”, please explain the issue in your comments.
Yes
8. Relevance of the topic to ISMIR: The topic of the paper is relevant to the ISMIR community. Note that submissions of novel music-related topics, tasks, and applications are highly encouraged. If you think that the paper has merit but does not exactly match the topics of ISMIR, please do not simply reject the paper but instead communicate this to the Program Committee Chairs. Please do not penalize the paper when the proposed method can also be applied to non-music domains if it is shown to be useful in music domains.
Strongly agree
9. Scholarly/scientific quality: The content is scientifically correct.
Agree
10. Please justify the previous choice (Required if "Strongly Disagree" or "Disagree" is chosen, otherwise write "n/a")
n/a
11. Novelty of the paper: The paper provides novel methods, applications, findings or results. Please do not narrowly view "novelty" as only new methods or theories. Papers proposing novel musical applications of existing methods from other research fields are considered novel at ISMIR conferences.
Agree
12. The paper provides all the necessary details or material to reproduce the results described in the paper. Keep in mind that ISMIR respects the diversity of academic disciplines, backgrounds, and approaches. Although ISMIR has a tradition of publishing open datasets and open-source projects to enhance the scientific reproducibility, ISMIR accepts submissions using proprietary datasets and implementations that are not sharable. Please do not simply reject the paper when proprietary datasets or implementations are used.
Agree
13. Pioneering proposals: This paper proposes a novel topic, task or application. Since this is intended to encourage brave new ideas and challenges, papers rated "Strongly Agree" and "Agree" can be highlighted, but please do not penalize papers rated "Disagree" or "Strongly Disagree". Keep in mind that it is often difficult to provide baseline comparisons for novel topics, tasks, or applications. If you think that the novelty is high but the evaluation is weak, please do not simply reject the paper but carefully assess the value of the paper for the community.
Disagree (Standard topic, task, or application)
14. Reusable insights: The paper provides reusable insights (i.e. the capacity to gain an accurate and deep understanding). Such insights may go beyond the scope of the paper, domain or application, in order to build up consistent knowledge across the MIR community.
Agree
15. Please explain your assessment of reusable insights in the paper.
The paper proposes an evaluation framework for measuring source preservation in music editing tasks with generative models. I'd say the evaluation framework itself is a reusable framework that will help us gain a more accurate and deep understanding of music editing models.
16. AI Usage Policy: The paper complies with the ISMIR 2026 AI Usage Policy for Authors. (Authors are welcome to use any tool they wish for preparing and writing the paper but they must ensure that all content is correct and original. Authors must declare any use of LLMs and AI tools if it's a part of core methodology (including literature review, related work, generated figures, tables or other illustrations). Using AI tools for editing purposes (e.g,. checking grammar and fixing typos) need not be declared. Please pay attention to any hallucinated text, figures, or references and flag them in the comments below)
Agree
17. Please justify the previous choice (Required if "Disagree" is chosen, otherwise write "n/a")
n/a
18. Write ONE line (in your own words) with the main take-home message from the paper.
The authors propose an evaluation framework for "music context preservation": it measures how well a music editing system (generative, audio domain) preserves the content (structure, tempo, rhythm, harmony, melody) of a piece of music while editing another desired aspect (like instrumentation or timbre) -- the authors validate their framework via an objective + listening validation study, and use their metrics to show how state of the art music editing models measure up in terms of the evaluation framework, highlighting the strengths and weaknesses of these models.
21. Potential to generate discourse: The paper will generate discourse at the ISMIR conference or have a large influence/impact on the future of the ISMIR community.
Disagree
22. Overall evaluation: Keep in mind that minor flaws can be corrected, and should not be a reason to reject a paper. Please familiarize yourself with the reviewer guidelines at https://ismir.net/reviewer-guidelines
Weak accept
23. Main review and comments for the authors. Please summarize strengths and weaknesses of the paper. It is essential that you justify the reason for the overall evaluation score in detail. Keep in mind that belittling or sarcastic comments are not appropriate.
The authors propose an evaluation framework for "music context preservation": it measures how well a music editing system (generative, audio domain) preserves the content (structure, tempo, rhythm, harmony, melody) of a piece of music while editing another desired aspect (like instrumentation or timbre) -- the authors validate their framework via an objective + listening validation study, and use their metrics to show how state of the art music editing models measure up in terms of the evaluation framework, highlighting the strengths and weaknesses of these models.

I recommend a weak accept. I like the initiative of the paper and think the metrics seem sound and well validated, both objectively and subjectively.

Strengths:
- I think the evaluation framework itself proposed by the author is comprehensive for the areas it does cover (harmony, melody, rhythm). It is well validated, with both an objective and listening validation, as well as using it to benchmarking existing models.

Weaknesses:

- The evaluation framework covers melodic/harmonic/rhythmic/tempo preservation. It does NOT cover timbre preservation. What if my edit is "change the melody, but preserve the timbre of my instruments"? This seems like a crucial edit instruction, and even if current models can't handle this very well, it would be good to design this metric and include it in the framework.

- I'm confused as to why the authors couldn't evaluate the "SOTA" models (MusicMagus, ZETA, Audio Prompt Adapter, etc.) using a common evaluation framework (same number of samples, edit instructions etc), at least in a limited manner, in order for the authors to be able to show what the evaluation framework looks like for *comparing* two models against each other.


- I don't know if I love the term "MCP" for "Music Context Preservation", as it will be confused with "Model Context Protocol", which is a way more prominent term in terms of LLM agents: https://en.wikipedia.org/wiki/Model_Context_Protocol.
Reviewer #3
Questions
2. I am an expert on the topic of the paper.
Disagree
3. The title and abstract reflect the content of the paper.
Agree
4. The paper discusses, cites and compares with all relevant related work
Agree
5. Please justify the previous choice (Required if “Strongly Disagree” or “Disagree” is chosen, otherwise write "n/a")
n/a
6. Readability and paper organization: The writing and language are clear and structured in a logical manner.
Strongly agree
7. The paper adheres to ISMIR 2026 submission guidelines (uses the ISMIR 2026 template, has at most 6 pages of technical content followed by “n” pages of references, AI usage declaration or ethical considerations, references are well formatted). If you selected “No”, please explain the issue in your comments.
Yes
8. Relevance of the topic to ISMIR: The topic of the paper is relevant to the ISMIR community. Note that submissions of novel music-related topics, tasks, and applications are highly encouraged. If you think that the paper has merit but does not exactly match the topics of ISMIR, please do not simply reject the paper but instead communicate this to the Program Committee Chairs. Please do not penalize the paper when the proposed method can also be applied to non-music domains if it is shown to be useful in music domains.
Strongly agree
9. Scholarly/scientific quality: The content is scientifically correct.
Strongly agree
10. Please justify the previous choice (Required if "Strongly Disagree" or "Disagree" is chosen, otherwise write "n/a")
n/a
11. Novelty of the paper: The paper provides novel methods, applications, findings or results. Please do not narrowly view "novelty" as only new methods or theories. Papers proposing novel musical applications of existing methods from other research fields are considered novel at ISMIR conferences.
Agree
12. The paper provides all the necessary details or material to reproduce the results described in the paper. Keep in mind that ISMIR respects the diversity of academic disciplines, backgrounds, and approaches. Although ISMIR has a tradition of publishing open datasets and open-source projects to enhance the scientific reproducibility, ISMIR accepts submissions using proprietary datasets and implementations that are not sharable. Please do not simply reject the paper when proprietary datasets or implementations are used.
Agree
13. Pioneering proposals: This paper proposes a novel topic, task or application. Since this is intended to encourage brave new ideas and challenges, papers rated "Strongly Agree" and "Agree" can be highlighted, but please do not penalize papers rated "Disagree" or "Strongly Disagree". Keep in mind that it is often difficult to provide baseline comparisons for novel topics, tasks, or applications. If you think that the novelty is high but the evaluation is weak, please do not simply reject the paper but carefully assess the value of the paper for the community.
Agree (Novel topic, task, or application)
14. Reusable insights: The paper provides reusable insights (i.e. the capacity to gain an accurate and deep understanding). Such insights may go beyond the scope of the paper, domain or application, in order to build up consistent knowledge across the MIR community.
Agree
15. Please explain your assessment of reusable insights in the paper.
The paper's most reusable contribution is the framing of music context preservation as a multi-facet problem (harmony, rhythm/meter, structure, melody/motif) with a per-facet diagnostic instead of a single aggregate score.
16. AI Usage Policy: The paper complies with the ISMIR 2026 AI Usage Policy for Authors. (Authors are welcome to use any tool they wish for preparing and writing the paper but they must ensure that all content is correct and original. Authors must declare any use of LLMs and AI tools if it's a part of core methodology (including literature review, related work, generated figures, tables or other illustrations). Using AI tools for editing purposes (e.g,. checking grammar and fixing typos) need not be declared. Please pay attention to any hallucinated text, figures, or references and flag them in the comments below)
Agree
17. Please justify the previous choice (Required if "Disagree" is chosen, otherwise write "n/a")
n/a
18. Write ONE line (in your own words) with the main take-home message from the paper.
A useful, well-motivated framework that reframes music-editing evaluation around preserving the attributes a user didn't ask to change, decomposed into per-facet metrics.
21. Potential to generate discourse: The paper will generate discourse at the ISMIR conference or have a large influence/impact on the future of the ISMIR community.
Agree
22. Overall evaluation: Keep in mind that minor flaws can be corrected, and should not be a reason to reject a paper. Please familiarize yourself with the reviewer guidelines at https://ismir.net/reviewer-guidelines
Weak accept
23. Main review and comments for the authors. Please summarize strengths and weaknesses of the paper. It is essential that you justify the reason for the overall evaluation score in detail. Keep in mind that belittling or sarcastic comments are not appropriate.
The paper proposes MuseCPEval, a multi-facet framework of 10 metrics that measures harmony, rhythm, structure, melody to study preservation of desired features in editing systems. Table 1 makes a convincing case that current systems evaluate preservation inconsistently. The metric choices seem sensible to me and the system case study is illustrative.

However, I found a few discrepancies in the supplement code submitted and the paper. It would be nice to resolve them and share a small dataset which the code be run against to reproduce and verify the results.

1. Structure facet can't run as shipped, and silently degrades. First of all, msaf is not in the requirements. If it was, requirements.txt pins scipy==1.13.0, but msaf (imported at structural_form.py:129) needs the removed scipy.inf. When the import fails, structural_score (structural_form.py:268-270) silently falls back to librosa_segments_fallback, which makes evenly-spaced segments with np.linspace.

2. ChromaDTWS: code reports the wrong variant. Paper Eq. 3 defines a raw DTW-path cosine. But main.py:313-315 reports best_shift_chroma_dtw_cosine: the transposition-invariant version.

3. Information Gain: formula vs. implementation mismatch. Eq. (6) normalizes by log₂K to [0,1]. The code (rhythm_meter.py:27) defines INFO_GAIN_MAX_BITS but never uses it; main.py:336-337 reports mir_eval's raw "Information gain" in bits instead.

4 Section 3.1 describes 50 Lakh MIDI samples under eight facet-spanning edits, but the supplied main.py downloads the MUSDB18 audio corpus instead (:15, :43) and applies a single edit, a fixed +6-semitone pitch shift (:102, :163-170). The other seven edits, the Lakh sample selection, and any MIDI rendering are absent from the supplement. The authors should release the actual validation code (sample IDs, rendering, and all eight edits) so the central empirical claim can be checked.