- Rating
- 2
- Confidence
- 2
- Informativeness
- 2
- Presentation
- 2
- Soundness
- 2
Read review
Paper summary
This paper proposes a framework that integrates hyperbolic representation learning with edl for improving OOD detection and uncertainty estimation. The authors explain that conventional EDL methods arenot suited to hyperbolic representations. The paper proposes three contributions. It introduces a constraint that constructs dirichlet evidence from the spatial components for avoiding geometric interference. It also develops a hyperbolic evidence inference that maps these representations into valid dirichlet concentration parameters.
Critical strengths
The paper shows competitive OOD detection performance against multiple benchmarks, with improvements on CIFAR-100 and long-tailed datasets. These improvements are supported by experiments across multiple random seeds and complementary metrics and evaluation metrics.
It also shows an interesting problem: how to construct dirichlet evidence from representations while consdiering their geometric structure. This proposed parameterization and LPEA illustrate an integrated architectural approach to connecting geometric representations with evidential uncertainty.
Critical weaknesses
The central argument and claim of the paper is based on the separation between radial information and class-discriminative semantics rather than a rigorous theoretical demonstration. Additionally the Lorentz time coordinate is deterministically related to the norm of the spatial components, it means excluding it does not eliminate the radial information from the evidence network. Then the proposed constraint does not establish the contribution of the semantic separation.
Also the main goal of the LPEA framework is to align evidential confidence with manifold geometry. But its loss only penalizes ground-truth evidence under a threshold, without enforcing a relationship between geometric distance and evidence strength. Furthermore, the experimental evaluation focuses on OOD detection metrics only which shows discrimination but do not establish the claimed of geometry–uncertainty alignment or probabilistic calibration.
Questions for authors
On of the paper contributions is that excluding the time radial coordinate prevents semantic contamination in evidence construction. However the time coordinate is completely related to the norm of the spatial components. How does the proposed spatial only parameterization guarantee that radial information does not influence the resulting of dirichlet evidence?
To what extent are the observed OOD detection improvements attributable to Lorentz representations rather than the additional evidence-alignment regularization? Could the authors provide a controlled comparison between Euclidean and Lorentz evidential models using the same backbone, evidence network, prototype mechanism, and training objective?
Table 3 of the paper shows reduced in distribution classification accuracy againts to baseline methods (under class imbalance). Why does the proposed method improve OOD detection but reduce classification accuracy? Whether OOD detection can be improved without reducing classification accuracy?
Are the improvements in OOD detection mainly because of using lorentz representations or the proposed evidence alignment method? Could the authors compare euclidean and lorentz models under the same experimental settings to show the contribution of each one?
5 comments
Responses to Reviewer KQ1M(Part 2)
Response to Weakness 2
We appreciate the reviewer’s valuable suggestions regarding alignment. We further explain the design rationale of and discuss why direct distance-to-evidence regression brings unique optimization difficulties in hyperbolic spaces, while our threshold-based formulation enables sound geometric alignment.
1. Considerations on Direct Distance-to-Evidence Regression in Hyperbolic Space: One simple scheme is to define a fixed functional mapping, e.g., letting Dirichlet evidence scale inversely with Euclidean or pseudo-Riemannian prototype distance. However, the Lorentz model of hyperbolic geometry has nonconstant negative curvature, and spatial volume expands exponentially rather than linearly. A rigid distance-evidence penalty applied across the entire manifold may trigger severe optimization conflicts:
- Metric Distortion: Distances in hyperbolic space are strongly distorted near manifold boundaries. A uniform penalty encourages the evidence network to follow flat Euclidean metric rules, weakening the hierarchical representational capacity that motivates adopting the Lorentz manifold.
- Intra-Class Dispersion vs. Inter-Class Separation: For complex or long-tailed data, discriminative samples of the same class do not converge to a single point; they form clusters with natural geometric spread. A strict distance-evidence constraint penalizes this reasonable intra-class variation, driving the model to overcompress features and impair closed-set decision boundaries.
2. How Achieves Robust Geometric-Evidential Alignment: Rather than enforcing an overly rigid global regression constraint, adopts a geometry-aware strategy based on large-deviation subsets ().
- Asymmetric Soft-Bounding: The threshold acts as a geometric boundary condition. It does not require a fixed proximity for valid evidence, but imposes a minimum acceptable evidence level for in-distribution samples that sit within reasonable geometric proximity to their class prototypes. By penalizing only samples that fall below this threshold, fixes abnormal evidence collapse and retains the continuous distance–evidence relationship over the rest of the manifold. This yields stable alignment compatible with the underlying topology.
Table 9: Ablation study for loss using CIFAR100 and CIFAR100‑LT () as ID datasets. (Note: Due to web response length limits, C100 denotes CIFAR-100, and C100LT denotes CIFAR-100-LT ().)
| Method | C100 | C100 | C100 | C100 | C100LT | C100LT | C100LT | C100LT |
|---|---|---|---|---|---|---|---|---|
| OOD | OOD | OOD | ID | OOD | OOD | OOD | ID | |
| FPR95$\downarrow$ | AUPR$\uparrow$ | AUROC$\uparrow$ | Acc$\uparrow$ | FPR95$\downarrow$ | AUPR$\uparrow$ | AUROC$\uparrow$ | Acc$\uparrow$ | |
| Ours(w/o ) | 40.29 ± 1.14 | 87.88 ± 0.52 | 91.52 ± 0.18 | 80.13 ± 0.17 | 56.07 ± 1.78 | 76.05 ± 1.04 | 84.92 ± 0.72 | 50.89 ± 0.46 |
| Ours | 36.50 ± 2.04 | 88.32 ± 0.69 | 92.24 ± 0.43 | 80.10 ± 0.13 | 48.25 ± 3.33 | 79.08 ± 1.03 | 88.36 ± 0.90 | 49.88 ± 0.62 |
3. Addressing the Evaluation of Geometric-Uncertainty Alignment: We agree that standard OOD metrics including FPR95 and AUROC evaluate discriminative performance instead of direct metric correlation. For evidential deep learning, a practical criterion for geometry–uncertainty alignment is whether evidence magnitude reliably captures epistemic uncertainty across regions with different manifold densities.
- Our controlled ablation between Ours (w/o ) and Ours as shown in part of Table 9 demonstrates that systematically improves uncertainty separation, particularly rescuing sparse tail classes from severe miscalibration. This confirms that our threshold-based design effectively governs evidential confidence across diverse manifold regions.
- We select the threshold-based design to preserve the hierarchical topology of the Lorentz manifold. That said, explicit functional regression under milder constraints is a promising complementary direction, which we reserve for future work (see the blue-highlighted text in Appendix E, Limitations and Future Work).
Responses to Reviewer KQ1M(Part 3)
Response to Question 2
We appreciate the reviewer raising this insightful point. To separate the OOD gains attributable to Lorentz geometry from those of the , we ran a controlled experiment. We reused the same evidence network, prototype design and training objective in standard Euclidean feature spaces, keeping all other experimental settings unchanged. Results on CIFAR100 and CIFAR100-LT () are shown below:
Table 9: Ablation study for loss using CIFAR100 and CIFAR100‑LT () as ID datasets. We additionally include a Euclidean counterpart equipped with its dedicated alignment loss for rigorous comparison. (Note: Due to web response length limits, C100 denotes CIFAR-100, and C100LT denotes CIFAR-100-LT ().)
| Method | C100 | C100 | C100 | C100 | C100LT | C100LT | C100LT | C100LT |
|---|---|---|---|---|---|---|---|---|
| OOD | OOD | OOD | ID | OOD | OOD | OOD | ID | |
| FPR95$\downarrow$ | AUPR$\uparrow$ | AUROC$\uparrow$ | Acc$\uparrow$ | FPR95$\downarrow$ | AUPR$\uparrow$ | AUROC$\uparrow$ | Acc$\uparrow$ | |
| EDL | 79.09 ± 0.85 | 71.51 ± 0.41 | 75.63 ± 0.51 | 80.38 ± 0.13 | 92.51 ± 0.62 | 58.90 ± 0.60 | 62.45 ± 0.71 | 58.13 ± 0.24 |
| EDL(w ) | 82.77 ± 0.03 | 62.87 ± 0.03 | 67.76 ± 0.03 | 79.65 ± 0.04 | 96.77 ± 0.05 | 47.64 ± 0.13 | 50.84 ± 0.21 | 53.81 ± 0.28 |
| Ours(w/o ) | 40.29 ± 1.14 | 87.88 ± 0.52 | 91.52 ± 0.18 | 80.13 ± 0.17 | 56.07 ± 1.78 | 76.05 ± 1.04 | 84.92 ± 0.72 | 50.89 ± 0.46 |
| Ours | 36.50 ± 2.04 | 88.32 ± 0.69 | 92.24 ± 0.43 | 80.10 ± 0.13 | 48.25 ± 3.33 | 79.08 ± 1.03 | 88.36 ± 0.90 | 49.88 ± 0.62 |
Note on EDL (w/ ): The extremely small standard deviations observed for EDL (w/ ) stem from rounding values originally in the magnitude (e.g., 0.0003463581241353642 for FPR95 and 0.00031084379783058355 for AUPR and 000343889819428845 for AUROC) to two decimal places, indicating a deterministic and highly stable convergence behavior under flat Euclidean geometry.
From these comparisons, we summarize the following observations:
- Lorentz geometry delivers the main performance gain. Switching from Euclidean space to the Lorentz manifold without alignment loss (EDL vs. Ours (w/o )) reduces OOD FPR95 by roughly half on CIFAR-100 (). This indicates the hyperbolic representation provides the core inductive bias for OOD detection.
- The regularization term alone does not help in Euclidean space. Adding directly to Euclidean EDL degrades performance. Flat Euclidean space lacks the exponential representational capacity of hyperbolic manifolds; imposing evidence constraints within this linear space compresses features and distorts decision boundaries.
- The two components work synergistically. brings further clear gains only when combined with Lorentz space. Negative curvature supplies enough representational capacity, so the alignment regularization can calibrate Dirichlet parameters without semantic collapse.
Taken together, the improvements arise from the synergy between Lorentz geometry and alignment regularization, where Lorentz representations form the basis. We have added these controlled results and the corresponding discussion to Table 9 in Appendix D.2 (see the blue-highlighted text).
Responses to Reviewer KQ1M(Part 1)
To facilitate the reviewer in identifying which weakness or question is being addressed, we have organized the index numbers as follows:
Critical Weaknesses: Weakness 1: The central argument ...
Weakness 2: Also the main goal of the LPEA ...
Questions For Authors:
Question 1: On of the paper contributions is ...
Question 2: To what extent are the ...
Question 3: Table 3 of the paper shows ....
Question 4: Are the improvements ....
Response to Weakness 1 & Question 1
We appreciate the reviewer raising this precise point. That is entirely correct: given the hyperboloid constraint z0=∥zs∥22+k, the time coordinate z0 and the spatial norm are mathematically tied. Our method does not "erase" radial information in an absolute, information-theoretic sense; rather, the core rationale is rooted in optimization dynamics and numerical stability.
Specifically, because z0 is strictly bounded below by k>0, it injects a permanent, non-zero scalar offset into the feature space. When the full [z0,zs] vector is fed directly into the evidence network—as tested in our Strategy-1 ablation in Table 4—this large positive scalar tends to dominate both the representations and the backpropagation gradients. Under skewed or heavy-tailed distributions like CIFAR100-LT, this common-mode offset overwhelms the fine-grained class semantics in zs, directly triggering the optimization pathologies that cause the sharp performance drop.
Restricting parameterization strictly to zs bypasses this numerical bias and gradient dominance while preserving the local manifold geometry needed for evidence estimation. We have updated Appendix B.1 in the revised manuscript to clarify this optimization-focused perspective (see the blue-highlighted text in Appendix B.1).