- Rating
- 3(+1)
- Confidence
- 3
- Informativeness
- 3
- Presentation
- 3
- Soundness
- 2
Read review
Paper summary
This paper introduces CONTRAIL, a counterfactual attribution framework for sparse Mixture-of-Experts models that separates the contribution of selected expert computation from the contribution of the router decision itself. For expert attribution, CONTRAIL treats the co-routed experts as a cooperative game and derives closed-form Shapley values from a quadratic approximation of the task loss, capturing both individual effects and co-routed interactions without enumerating all coalitions. For router attribution, it introduces the Coalition-Consistent Gate Estimator (CCGE), which evaluates a mass-conserving swap between the lowest-logit selected expert and the highest-logit excluded expert. The method is validated on four instruction-tuned sparse MoE models across language modeling, mathematical reasoning, and knowledge-intensive tasks. Experiments on expert removal, cross-task transfer, router interventions, interaction ablation, and sparse expert fine-tuning show that CONTRAIL identifies more consequential experts and routing decisions than routing frequency and other attribution baselines, while also revealing task-dependent expert importance and useful targets for parameter-efficient adaptation.
Critical strengths
Strength 1: The paper clearly separates route-conditioned expert attribution from router attribution, which avoids mixing expert computation with routing changes.
Strength 2: CONTRAIL provides a closed-form Shapley formulation that captures both individual expert effects and co-routed interaction without enumerating all expert coalitions.
Strength 3: The experiments are comprehensive, covering four sparse MoE models, multiple tasks, expert removal, router interventions, cross-task analysis, and sparse fine-tuning.
Critical weaknesses
Weakness 1: CONTRAIL claims to model co-routed interaction through the term in Eq. (3), but the implementation replaces with the rank-one surrogate in Eq. (10). This makes every pairwise interaction reduce to in Eqs. (70)–(73). In other words, two experts are judged to interact only through how their outputs align with the same gradient vector . Any interaction that is reflected in other curvature directions is lost. This is a strong restriction, and the paper does not show how close this rank-one interaction is to the interaction obtained with the true Hessian.
Weakness 2: Eq. (62) gives a bound on the difference between the exact nonlinear expert value and the practical CONTRAIL score, but the paper does not show whether this approximation keeps the same expert ranking. Figure 6 only shows that the ranking becomes stable as more probe examples are used. It does not tell us whether the resulting ranking matches the exact nonlinear expert game. Since the Top 1% and Top 3% interventions depend entirely on which experts appear at the top of the ranking, it would be important to know whether approximation error changes these top-ranked sets.
Weakness 3: Under the empirical-Fisher-style surrogate, Eq. (74) can be written as . For a fixed token and MoE layer, the second factor is shared by all active experts, so the full contribution does not introduce an expert-specific interaction correction relative to the direct term at that token. Its effect instead comes from changing the weight of different token-level contributions before corpus-level aggregation. This makes it unclear whether the improvement attributed to co-routed interaction in RQ4 is actually due to expert–expert interaction or to token-level reweighting induced by .
Questions for authors
Question 1: How accurate is the rank-one curvature surrogate in Eq. (10) for estimating the co-routed interaction term in Eq. (3)? Can you report the approximation error?
Question 2: Does the practical CONTRAIL score preserve the expert ranking of the exact nonlinear expert game? Can you report the rank correlation or Top- overlap between the practical scores and exact nonlinear coalition-based scores?
Question 3: How often does the interaction term in Eq. (74) actually change the ordering of active experts for a given token and MoE layer? Can you report the token-level ranking difference between the full contribution score and the direct or self-only score, and then show how much additional change is introduced by corpus-level aggregation?
4 comments
W2 / Q2. Agreement with nonlinear Shapley rankings and Top-K sets
The probe-prefix experiment addresses sampling stability. To assess fidelity to the nonlinear game, we compare rankings on identical tokens, with inactive-expert zeros and the same aggregation rule. A score-error bound preserves the Top- set when the exact selection-boundary gap exceeds ; the direct comparison below evaluates the resulting selections.
The sample covers all 64 OLMoE experts and all 32 gpt-oss experts in each layer/task stratum. We use raw signed corpus scores, average ranks for correlation ties, and fixed expert IDs to resolve selection-boundary ties. The nominal 1% budget selects per layer in both models; the nominal 3% deletion budget selects , including the gpt-oss configuration override. The ranking comparison below reproduces Table 20 in Appendix H.3.
| Practical score vs. nonlinear reference | Spearman | Kendall | overlap | overlap |
|---|---|---|---|---|
| Within-token active-expert ranking | 0.82 | 0.68 | — | — |
| Layer-local corpus expert ranking | 0.89 | 0.74 | 13/16 = 81.25% | 27/32 = 84.375% |
Correlations are averaged across units or strata. Overlap is ; the fractions aggregate the 16 layer/task/model strata.
We also compare removal effects on 128 additional packed blocks per model/task, selecting experts in each of the four sampled layers. Average NLL increases are 0.061 for Direct, 0.071 for HVP, 0.068 for rank-one Full, and 0.074 nats/target token for nonlinear-Shapley selection. Together, these results indicate useful ranking agreement on the evaluated subset, with remaining differences at the selection boundary. The revised manuscript specifies the corpus aggregation and selection rules in Section 4.7 and Appendices C.3 and H.3. As clarified in Appendix H.3, corpus scores already include routing frequency through ; no additional routing-frequency multiplier is applied in this comparison.
Expert-removal results (Table 21, Appendix H.3).
| Expert selector | Mean NLL increase (nats/target token) |
|---|---|
| Direct | 0.061 |
| HVP | 0.071 |
| Rank-one Full | 0.068 |
| Nonlinear-Shapley selection | 0.074 |
W3 / Q3. Token reweighting and changes in expert ordering
Your algebraic observation is correct. Define , , and . The Full score is . At a fixed token/layer, positive preserves every strict Full/Direct ordering, negative reverses it, and zero collapses the full scores to a tie.
The comparison with Self-only, , follows from the pairwise differences. Suppressing ,
=(s_e-s_m)(-1+\tfrac12s_\Sigma),\qquad \phi_e^{\mathrm{self}}-\phi_m^{\mathrm{self}} =(s_e-s_m)(-1+\tfrac12(s_e+s_m)).$$ A strict Full/Self reversal occurs when the two parenthesized factors have opposite signs and $s_e\ne s_m$. After aggregation, $\widehat\Phi_e=\frac1N\sum_i w_i d_{i,e}$, so **varying token weights can change corpus rankings even when every weight is positive**. Across the same 1,024 units, 976 weights are positive (95.31%), 40 are negative (3.91%), and 8 are near zero (0.78%), using $|w_i|\le10^{-6}$. Each model contributes 512 units: 488 positive, 20 negative, and 4 near zero. The eight **near-zero units** are excluded from the pairwise ranking comparison; the remaining score pairs are non-tied. Full/Direct reverses 680 of 17,272 pairs (3.94%), and Full/Self reverses 1,164 pairs (6.74%). The corpus-ranking comparisons below are reported in **Table 22** (Appendix H.4). | Corpus-level comparison | Spearman $\rho$ | Kendall $\tau_b$ | $K=1$ overlap | $K=2$ overlap | | ----------------------- | -------------- | --------------- | -------------- | -------------- | | Full / Direct | 0.92 | 0.78 | 11/16 = 68.75% | 24/32 = 75.00% | | Full / Self-only | 0.90 | 0.76 | 12/16 = 75.00% | 26/32 = 81.25% | These results show how limited within-token reversals can accompany larger changes in the corpus-level selected sets. **Weighted Direct and Full are algebraically identical**, giving correlation and overlap of 1 wherever defined. Section 4.5 now interprets RQ4 as an ablation of this `token-reweighting correction`. Its existing results—greater removal damage for Full in 20 of 36 cells, with two ties at the displayed precision—show that the correction's utility varies across settings.Thank you for the additional experiments. The ranking correlations, Top-k overlaps, and expert-removal results address my second concern. The analysis of the shared correction factor also confirms my observation in W3 and explains why the main changes in expert ranking occur after corpus-level aggregation.
For W1, the new results help clarify the accuracy of the rank-one approximation. The correlation of 0.49 with nonlinear interaction residuals is still relatively low, suggesting that the approximation misses a substantial part of the interaction information. I think this remains a limitation of the method, although the authors have now provided the evidence needed to assess it.
The revisions have addressed most of my concerns, and I have adjusted my rating accordingly.
We sincerely thank the reviewer for the thoughtful follow-up and careful reassessment of our revised manuscript. We are glad that the additional experiments and revisions have addressed most of your concerns, and we appreciate your updated evaluation.
Your constructive feedback has helped us improve both the empirical analysis and the clarity of our presentation. Thank you again for your time and valuable suggestions.
Response to Reviewer 1E3Y
Thank you for the careful analysis of the practical curvature approximation and the shared token-level factor. These observations help clarify what the implemented correction contributes. The corresponding revisions have been incorporated into the revised manuscript PDF and highlighted in red. The updates clarify the practical estimator, token reweighting, and ranking fidelity in Sections 3.3, 4.5, and 4.7, with detailed comparisons in Appendices H.2–H.4.
The
gradient outer productis a directional curvature model. With si,e=ui⊤δi,e, its pair term is ψi,em=si,esi,m, and its error relative to the quadratic interaction is ψi,em−ψi,em(2)=δi,e⊤(uiui⊤−Hi)δi,m. This expression identifies the curvature information omitted by the approximation.We evaluate these contractions on 1,024 token/layer units from four layers each of OLMoE and gpt-oss, across WT-2 and GSM8K. Rank-one pair terms have normalized error 0.62 and Spearman correlation 0.60 against local HVP contractions. Against nonlinear joint-removal residuals, the corresponding values are 0.77 and 0.49. HVP pair terms themselves have error 0.41 and correlation 0.78 against the nonlinear residuals, showing that finite-removal approximation also contributes to the discrepancy.
Expert-value and pairwise-interaction fidelity (Table 7, Section 4.7). The comparison on these attribution units is reproduced below.
Table 19 in Appendix H.2 reports sign agreement of 76% with HVP pair terms and 69% with nonlinear residuals, with 95% intervals of [72%, 80%] and [64%, 74%]. The respective majority-sign baselines are 57% and 56%. We exclude near-zero references from this statistic using a threshold of 1% of the stratum median absolute nonzero interaction; these account for approximately 12% of pairs.
The comparison supports partial recovery of interaction signs and ordering, with appreciable error in magnitude. Section 4.7 and Appendix H.2 now describe the pair terms as rank-one interaction approximations and report their accuracy at this level.