- Rating
- 2
- Confidence
- 2
- Informativeness
- 2
- Presentation
- 2
- Soundness
- 2
Read review
Paper summary
The paper proposes a framework for learning group decision-making over a sequence of episodes, aiming to minimize the maximum dissatisfaction across participants. It utilizes contextual bandits and bilinear modeling to represent participants' utility functions, along with a pretrained encoder, potentially an LLM-based encoder, to generate embeddings of the decision options. The proposed algorithm jointly learns a shared projection matrix and individual participant preference vectors through alternating minimization, using scalar utility feedback collected across episodes. The authors provide a theoretical cumulative pseudoregret bound that depends on the spectral dimension of the representations rather than the ambient encoder dimension. They also present experimental results evaluating the proposed framework using synthetic simulations and pretrained text encoders.
Critical strengths
-
The main strength is the elegant problem formulation and representation-guided preference learning. The paper presents a formulation of multi-episode group decision-making, utilizing pretrained text encoders, potentially LLM-based, to represent decision options. Furthermore, the framework requires participants to provide only scalar utility feedback for each presented plan, without explicitly specifying their preference functions or ranking all possible options, adding to the simplicity and elegance of the proposed approach.
-
The suggested algorithm of the joint learning and preference generalization is also very strong. By jointly learning a shared projection matrix and participant-specific preference vectors, the proposed method aims to identify preference-relevant similarities between options and generalize to unseen decisions across episodes.
-
Theoretical analysis: The paper provides a theoretical cumulative pseudoregret bound that depends on the spectral dimension of the encoder representations rather than their ambient dimension. The theoretical analysis and supporting proofs are valuable contributions that help explain the potential advantages of low-dimensional representations.
-
The results section is very informative and contains very extensive ablation studies.
Critical weaknesses
-
Lack of real-world validation of the fundamental modeling assumptions: The proposed framework relies on strong assumptions regarding participants' utility functions, particularly that preferences can be represented through a shared low-rank bilinear model and remain stable across episodes. However, all experiments use synthetic participant preferences and utility functions, including those involving real pretrained encoders. Therefore, the experiments do not establish whether the proposed model accurately represents real human preferences or whether pretrained language representations capture the features relevant to actual decision-making. Those assumptions form the foundation of the proposed framework, thus their lack of validation on real human preference data significantly limits the practical conclusions that can be drawn.
-
Theoretical guarantees rely on strong, unverified assumptions and an idealized algorithm: Although the paper provides a cumulative pseudoregret bound, several critical properties required for the guarantee are assumed rather than established. In particular, assumption A3(i) requires sufficient exploration of the relevant representation space, without proving that the proposed LCB-minimax strategy ensures this condition. Assumption A4(ii) directly assumes that the alternating-minimization procedure contracts toward the true parameters, rather than proving this property for the proposed algorithm. Furthermore, Theorem 1 applies to an idealized coordinator that differs from the implemented Algorithm 1 in its optimization procedure, confidence bounds, and projection estimation.
-
Insufficient comparison with competitive baseline methods: While as mentioned in the strengths the ablations study are extensive and compares the proposed approach against variations of the same coordinator, such as random projections, PCA-based representations, there is no clear baseline of other methods or clear SOTA to compare to such as bilinear bandit, contextual bandit, or preference-learning methods. Additionally, the quality of the proposed combinatorial minimax optimization is not adequately evaluated against stronger global optimization approaches or exact solutions on smaller instances. Without these comparisons, it is difficult to assess whether the proposed method offers a substantial improvement over existing alternatives.
-
Questionable exploration–exploitation behavior: In the synthetic experiments, the proposed joint-learning algorithm substantially outperforms the baseline provided with the true projection matrix . The authors attribute this result to the additional exploration induced by repeatedly updating , whereas the fixed-true- baseline reportedly becomes overconfident and exploits prematurely. However, this explanation raises concerns about the exploration–exploitation mechanism itself. An algorithm given the correct representation should not inherently be disadvantaged by this additional information. The results suggest that the observed improvement may partly arise from inadequate exploration or uncertainty calibration in the fixed-representation baseline, rather than from the superiority of the proposed representation-learning procedure. A comparison with a properly calibrated exploration strategy using the true \\ would help clarify this issue.
-
Limited expressiveness of the additive utility model: The proposed utility representation constructs each plan's features by summing the embeddings of its selected options. Consequently, the modeled utility is additive across decisions and does not explicitly capture interactions or dependencies between different choices. For example, a participant might individually prefer remote work and frequent in-person meetings but dislike their combination. Such interactions are common in realistic group decision-making and cannot be represented directly by the proposed bilinear additive model. The paper does not investigate the effect of this limitation or evaluate more expressive alternatives.
-
Lack of robustness to strategic preference manipulation: The proposed minimax objective relies on participants providing meaningful and comparable scalar utility ratings. However, strategic participants could exaggerate their preferences or dissatisfaction to disproportionately influence the selected plan. In particular, the minimax objective may prioritize a participant reporting exceptionally large utility differences over the remaining participants. The paper does not investigate strategic reporting, utility-scale normalization, incentive compatibility, or robustness to adversarial participants. This is particularly relevant given the paper's emphasis on fair group decision-making.
-
Confounded experimental validation of effective-rank scaling: In Section 4.2.2, the authors investigate the relationship between effective representation rank and performance by increasing the Gaussian noise added to the synthetic embeddings. However, this intervention simultaneously increases the effective rank and deteriorates the quality of the underlying representation. Therefore, the observed increase in regret cannot be attributed specifically to effective rank, as it may simply result from a lower signal-to-noise ratio. Furthermore, the experiment measures stable rank , whereas the theoretical result depends on the spectral dimension . A controlled experiment varying representation rank while preserving signal quality would provide more convincing evidence for the theoretical claims.
-
Limited justification for using large LLM encoders: Although the paper motivates its approach through LLM-mediated group decision-making, the actual algorithm uses pretrained encoders only to extract representations of textual options, while participants provide scalar utility feedback. Moreover, the relatively small MiniLM sentence encoder outperforms all three larger LLMs in the reported experiments. While this demonstrates that the framework is representation-agnostic, it also raises questions about the practical benefits of using large LLM encoders instead of smaller specialized encoders or alternative learned representations.
-
The abstract seems to contain an error "46.57–146" should be
Questions for authors
-
Validation on real human preferences: Could the authors evaluate the proposed framework using real human preference data? In particular, how well do the assumptions of low-rank bilinear utilities, additive preferences, and stable participant preferences hold in realistic group decision-making scenarios?
-
Theoretical assumptions and practical guarantees: Can the authors establish that the proposed coordinator satisfies the exploration condition A3(i) and the alternating-minimization contraction condition A4(ii), rather than assuming these properties?
-
Comparison with established baselines: Could the authors compare their approach against existing bilinear bandit, contextual bandit, or preference-learning algorithms? Additionally, could they evaluate the quality of their minimax optimization against stronger global optimization methods or exact solutions on smaller instances?
-
Exploration–exploitation behavior: Why does the proposed joint-learning algorithm substantially outperform the baseline using the true projection matrix ? Could the authors investigate whether this result is caused by insufficient exploration or inaccurate uncertainty calibration in the fixed- \\ baseline, and compare against a properly calibrated exploration strategy?
-
Strategic preference manipulation: How would the proposed coordinator handle participants who strategically exaggerate their utility ratings to influence the minimax objective?
-
Effective-rank experiment: In Section 4.2.2, increasing Gaussian noise simultaneously increases effective rank and reduces representation quality. Could they provide a controlled experiment that varies effective rank while preserving signal quality?
-
Benefits of large language model encoders and natural-language feedback: Given that MiniLM outperforms the larger LLM encoders in the reported experiments, under what circumstances do the authors expect LLM-based representations to provide an advantage over smaller specialized encoders or alternative learned representations? Furthermore, could the framework be extended to utilize LLMs to interpret and quantify natural-language feedback from participants, rather than relying exclusively on scalar utility ratings? For example, participants may express conditional preferences ("I support remote work, provided that team meetings remain in person"), conflicting priorities ("I value flexibility, but not at the expense of collaboration"), or qualitative concerns that cannot be adequately captured by a single numerical rating. Such an extension could better exploit the capabilities of LLMs and improve the applicability of the framework to realistic group decision-making scenarios.
2 comments
Dear Reviewer,
With the discussion period closing soon, we wanted to kindly follow up and ask whether our rebuttal and revisions have adequately addressed your concerns. We would be happy to provide any further clarification if needed.
Thank you for your time and consideration.
We thank the reviewer for describing the formulation as elegant, the joint learning and preference-generalization algorithm as strong, the theoretical analysis as a valuable contribution, and the ablation studies as extensive and informative. We also appreciate the comprehensive nine-point review. We address each item in order next. We have marked all updates in blue text within the manuscript.
W1/Q1. It is difficult to conduct a real-human experiment in the rebuttal window, but we flag this as the central empirical gap in the combined Limitations/Future Work paragraph.
W2/Q2. The reviewer is correct that A3(i) and A4(ii) are stated conditionally and not established for Algorithm 1. We have made three changes: (i) we split A4(ii) into A4(ii-W) (pooled, dominant term) and A4(ii-φ) (per-user, scaling only in mi), in order to preserve the dominant term; (ii) we measure sinΘ(W^t,Wfit) and the smallest top-kh Gram eigenvalue along Algorithm 1's trajectory on the simulator and all four real encoders, showing both stabilize as A3(i) and A4(ii) require (Figure 2, appendix); (iii) we add Proposition 2 in Appendix B.1.2 giving a trajectory-independent excitation lower bound for a forced-exploration variant of Algorithm 1 with ηt∼t−1/3 uniform mixing, providing a theoretical anchor for the conditional A3(i).
W3/Q3. We add a GLM-dueling bandit (Faury et al. 2020, adapted) and a bilinear bandit (ESTR, Jun et al. 2019) on the simulator, matched in interaction budget (Appendix C.5). GLM-dueling is dominated by joint (Gtotal=97.84±6.39 vs joint's 22.06±1.47), reflecting its halved per-round information budget and lack of cross-user pooling. Bilinear-bandit ESTR matches joint within one SE on both Gtotal (18.24±1.11) and Rtotal (916±40 vs 877±64); this is because ESTR is designed exactly for the bilinear reward structure the simulator instantiates. Our contribution is the multi-episode group-coordination setting (per-user ϕi, cross-episode option drift, plug-and-play hyperparameters).
W4/Q4. The reviewer's concern was well taken: we had not shown that the joint vs. fixed-true-W⋆ gap survives proper exploration calibration. We sweep an inflation factor $\kappa\in\lbrace 0.5,1,2,4,8,16,32,64,128\rbrace $ on fixed-true-W⋆'s βt (range extended after no U-shape appeared at small κ). No value of κ closes the gap (best setting at κ∈{16,32} yields Gtotal≈44 vs. joint's 22.06). We now interpret the improvement as cross-episode W-refit injecting plan-space feature diversity through the refit trajectory. We added Figure 3 and a corresponding interpretive paragraph.
W5/Q5. We agree this was missing. A paragraph has been added to Limitations.
W6/Q6. The reviewer is correct that the σh sweep changes both rH and signal-to-noise. We have added a controlled sweep that varies the singular-value profile of W⋆ (specifically, the kh-th eigenvalue) while fixing ∥W⋆∥F and the misspecification residual ε=0. Appendix C.4 shows Gtotal and Rtotal grow monotonically with kh at log-log slopes 0.14 and 0.28 — below the theoretical 0.5 reference, consistent with Theorem 1 providing a valid upper bound that simulator performance clears comfortably. We retain the original σh sweep as a joint rank-and-noise stress test and clarify its interpretation.
W7. Addressed by W6/Q6 above.
W8/Q7. The MiniLM-beats-8B finding is deliberately central to the paper's empirical claim: our method is representation-agnostic and benefits most from encoders whose top-kh subspace aligns well with the utility-carrying features, regardless of model size. This is consistent with the theorem's kh-dependence. We also acknowledge an extraction-method confound in the comparison: MiniLM uses mean-pooled sentence embeddings (its canonical extraction) while the three 8B LLMs use last-token last-layer activations (the standard extraction for causal LLMs). A controlled ablation applying mean-pooling to the 8B hidden states would separate these contributions; we have added this caveat to Limitations and defer the ablation to future work. The reviewer's natural-language-feedback suggestion is excellent and we have added it to Future Work as a natural methodological extension.
W9. The "46.57–146" is the range 46.57–145.56 written in shortened form; we have expanded it to 46.57–145.56 in the revision.