- Rating
- 3(+1)
- Confidence
- 3
- Informativeness
- 3(+1)
- Presentation
- 1
- Soundness
- 2
Read review
Paper summary
The paper studies how replacing standard additive residual connections with Attention Residuals affects token dynamics. The authors introduce a continuous-depth model that accounts for the aggregation of previous layer outputs, and derive coupled equations for token directions and lengths. Under additional assumptions on the attention matrices and history weights, they show that almost every initial configuration converges directionally to at most two clusters, corresponding to the two opposite directions of the leading eigenvector of the value matrix.
Critical strengths
The motivation for the paper - studying how depth-wise aggregation of attention residuals affects token dynamics is a well-motivated direction. I think the derivation and setting the authors consider could be used to draw important insights.
Critical weaknesses
- I found the paper hard to read, both the main text but especially the proofs. I left some notable cases in the questions but this is true for most of the paper, it is very hard to follow the mathematical arguments that are presented densely.
- The contributions the author draw from their work is the analysis of the derivation of the dynamics, the authors do not draw any insight on what is the difference between Attention Residuals and the standard token dynamics analysis, this should be highlighted as it is the main motivation for the paper.
Questions for authors
- Could the authors either rephrase this paragraph or explain what they wanted to state here? (line 80) "The almost-everywhere conclusion concerns this raw initial state space, including the radii; it does not by itself give the same assertion for every fixed tuple of initial radii. The results do not claim a complete classification for arbitrary trained Transformers."
- Please explicitly refer to as the score matrix when introducing it, if this terminology is used later
- Please rewrite the second paragraph under assumption 1, it is very hard to follow (line 211).
- Most of the proofs are very hard to follow, for example the proof of Lemma 5.5 is described in large consecutive of text describing mathematical derivation but not showing it.
- The first two paragraphs of section 5 are hard to follow, it seems the result is introduced twice.
- Did you try to develop numerical simulations that show your results hold in practice?
6 comments
Response to Reviewer uQQM (2/2)
[Q2] Please explicitly refer to as the score matrix when introducing it, if this terminology is used later.
Answer. In the revised manuscript, we explicitly define as the score matrix in Section 3.2 and consistently use this terminology throughout the paper.
[Q3] Please rewrite the second paragraph under assumption 1, it is very hard to follow (line 211).
Answer. We have revised the corresponding paragraph in the manuscript as follows:
Requiring reads . Here is the condition number of , and measures how far the score matrix is from . In words, the attention scores must be large compared with the spread of 's eigenvalues, and more so for longer sequences (larger ) or for score matrices further from isotropic (larger ). In the simplest case (so ), the condition is . The regime is not empty: with , and , we get and .
[Q4] Most of the proofs are very hard to follow, for example the proof of Lemma 5.5 is described in large consecutive of text describing mathematical derivation but not showing it.
Answer. Thanks for your comments. We have revised the part of the proof that follows the linearization in the revision, so that every step is now argued explicitly:
- the equation near the non-polar equilibrium set is written in the form of Lemma 5.4, with the explicit matrices and ;
- the derivatives of the remainders are bounded explicitly, and a path argument gives the Lipschitz constant that Lemma 5.4 requires;
- for the measure-zero conclusion, the states that stay near the equilibrium set form a Borel null set, by Lemma 5.4 and Tonelli's theorem. The flow map is a local diffeomorphism by Liouville's formula, so the corresponding initial values also form a null set.
We have also rewritten the second half of the proof of Lemma 5.3 in the same way. It now proves explicitly that:
- the potential converges along the trajectory, because after some time it can cross a non-critical level only upwards;
- every limit point of the token direction is a critical point, and the limit set lies in a single connected component of the critical set;
- in the non-polar case, the radius converges to , so the raw token approaches .
[Q5] The first two paragraphs of section 5 are hard to follow, it seems the result is introduced twice.
Answer. Thanks for pointing this out. In our revision, we have deleted the second paragraph, so the result is now introduced only once.
[Q6] Did you try to develop numerical simulations that show your results hold in practice?
Answer. Following your suggestion, we have added Appendix E to the revised manuscript, which presents the following numerical simulations.
- An actual Attention Residual Transformer (Appendix E.1). We use a PyTorch implementation without modifying its architecture, with 1000 layers, and random tokens.
- With parameter values satisfying our assumptions (Fig. 1), the tokens converge to the two poles predicted by Thm. 5.1, and the split between the poles stays the same at every depth. With the standard residual, every token is within of its pole at layer 1000. Full Attention Residuals are slower, as their history rate predicts.
- With Assumption 1c violated (Fig. 3, Appendix E.2), the tokens still reach , but in four of the six runs almost all of them gather at one pole.
- From the random initialization (Fig. 2), which lies outside our assumptions, the tokens instead gather on one side of a single direction.
- Pretrained models (Appendix E.3, Fig. 4). Three public Attention Residual language models, which also lie outside our assumptions, show no two-pole structure.
We hope we have cleared your concerns about our work. We have also revised our manuscript according to your comments, and we would appreciate it if we could get your further feedback at your earliest convenience.
References
- [1] Kimi Team. Attention Residuals. arXiv:2603.15031, 2026.
- [2] B. Geshkovski, C. Letrouit, Y. Polyanskiy, P. Rigollet. The emergence of clusters in self-attention dynamics. In NeurIPS, 2023.
- [3] N. Karagodin, Y. Polyanskiy, P. Rigollet. Clustering in causal attention masking. In NeurIPS, 2024.
Any Questions from Reviewer uQQM on Our Rebuttal?
We would like to thank the reviewer again for your thoughtful review and valuable feedback.
We would appreciate it if you could let us know if our responses have addressed your concerns and whether you still have any other questions about our rebuttal.
We would be happy to do any follow-up discussion or address any additional comments.
Response to Reviewer uQQM (1/2)
Thank you for your thoughtful review and constructive feedback. We have revised our manuscript in response to your suggestions, with all changes highlighted in blue. Below, we address each of your concerns in detail.
[W1] I found the paper hard to read, both the main text and especially the proofs. I left some notable cases in the questions, but this is true for most of the paper; it is very hard to follow the mathematical arguments that are presented densely.
Answer. Thanks for your comments. We have elaborated our results and our mathematical arguments throughout the revision; the specific cases you raised are answered in Q1–Q5 below.
[W2] The contributions the author draw from their work is the analysis of the derivation of the dynamics, the authors do not draw any insight on what is the difference between Attention Residuals and the standard token dynamics analysis, this should be highlighted as it is the main motivation for the paper.
Answer. Thanks for your suggestion. In our work, we analyze Attention Residuals as proposed by the Kimi Team [1]. Prior work models residual Transformers in continuous depth by x˙i=Pxi∑jpijVxj on the sphere [2, 3]. Our dynamics (Thm. 3.4) differ from this model in three ways.
The limiting directions ±e, set by the value matrix, are shared with residual Transformers. Whether tokens reach them, how fast, and with which norms is specific to Attention Residuals. We added this comparison to the revised manuscript.
[Q1] Could the authors either rephrase this paragraph or explain what they wanted to state here? (line 80) "The almost-everywhere conclusion concerns this raw initial state space, including the radii; it does not by itself give the same assertion for every fixed tuple of initial radii. The results do not claim a complete classification for arbitrary trained Transformers."
Answer. Thanks for pointing this out. Here, "raw" means the original token vector ui∈Rd, i.e., its direction and its length. Following your suggestion, we have rewritten that paragraph in our revision as follows: