- Rating
- 3(+1)
- Confidence
- 3
- Informativeness
- 3
- Presentation
- 2
- Soundness
- 2
Read review
Paper summary
The paper proposes a conditional image generator built around a continuous dynamical system with Ising- or Kuramoto-style latent variables. The latent dynamics are designed to decrease an energy function, and class conditioning is introduced locally through class-dependent pairwise couplings rather than by broadcasting a global conditioning vector. A convolutional decoder maps the final latent state to an image. The paper reports CIFAR-10 generation results, ablations over conditioning choices, robustness to quantization and noise, and projected energy/latency figures for a possible mixed-signal implementation.
At the model level, the main claim is that local conditioning in the latent dynamics contributes more than placing the condition only in the decoder or reservoir. At the hardware level, the paper argues that the dynamical-system structure could support substantially more efficient generation than a GPU implementation.
Critical strengths
-
The co-design idea is coherent: conditioning is part of the physical interaction law rather than an additional digital pathway. The comparison against reservoir-only and decoder-only conditioning supports the claim that the latent dynamics are doing useful conditional computation.
-
The paper does more robustness analysis than is common in this area. Quantization, device noise, coupling perturbations, and several architectural ablations are reported, so the model-level behavior is not supported by a single idealized experiment.
Critical weaknesses
-
The headline hardware numbers combine evidence from different sources and levels of abstraction: a 64-spin 65 nm post-layout block, an extrapolation to 4096 spins, and peak efficiency figures from third-party compute-in-memory designs. I could not reconstruct a single end-to-end system corresponding to the reported 5.9 microjoules and 57.4 microseconds. Data conversion, memory movement, control, I/O, and decoder costs are not accounted for consistently, and the GPU baseline is not specified well enough to reproduce. This makes the strongest efficiency claim premature.
-
The scaling argument does not yet match the dense model that is evaluated. A 4096-spin dense pairwise system contains roughly 8.39 million couplers, whereas the most favorable energy case assumes bounded degree. There is no area, routing, or parasitic analysis for the dense array. Similarly, the Lyapunov argument holds for the ideal symmetric continuous dynamics, but the robustness experiments do not show that monotone energy descent survives the saturation, asymmetry, discretization, and noise expected in the proposed implementation.
Questions for authors
-
Can the authors provide one auditable end-to-end cost table for the proposed generator, separating oscillator/dynamics, coupling storage and access, converters, control, decoder, and I/O? Please use the same inclusion rules for the hardware and GPU baselines.
-
What physical topology realizes the dense 4096-spin interaction matrix, and what are the resulting area, wire, and bandwidth costs? If the intended topology is sparse, what generation result is obtained with that exact sparsity?
-
Does the implemented noisy, quantized, and asymmetric system still exhibit monotone energy descent? A trajectory-level measurement would directly connect the hardware study to the theoretical argument.
4 comments
Response to Reviewer tVwc (Part 2/2)
W2-a / Q2. Scaling to larger systems. Physical topology of the interaction. Area, routing, and parasitic analysis.
Scaling to larger systems. Our goal is not to replace GPU-native generators by scaling the dense design, but to show that hardware can generate images with its own physics. To scale further, we are finalizing a banded design for tape-out, in which each spin connects only to a fixed number of neighbors. With band width K = 32, each spin couples to its 2K = 64 nearest neighbors, so the design needs ~64x fewer couplers than the dense one (128x with K = 16). The coupler count and power then grow linearly with n instead of quadratically. The results below show a modest FID cost against the dense design, and both banded models still can have better performance than the baseline.
| Coupling | DS couplers | Clean-FID (3 seeds) | Mean ± std |
|---|---|---|---|
| Banded, K = 32 | 0.131M | 7.81 / 6.71 / 7.24 | 7.25 ± 0.55 |
| Banded, K = 16 | 0.065M | 7.28 / 9.42 / 7.58 | 8.09 ± 1.16 |
Physical topology of the interaction. Area, routing, and parasitic analysis. The dense interaction matrix is implemented using a 4 * 4 tiled architecture, with each tile containing 1024 * 1024 coupling elements, corresponding to a total coupling area of 67.09 mm². Unity-gain voltage followers distribute spin voltages between tiles, while grounded metal shielding minimizes crosstalk. Calibre PEX extraction shows 0.1 fF inter-wire coupling and 9.2 fF ground parasitic capacitance per 500 µm. The four-stage voltage follower chain is designed for a 650-MHz end-to-end bandwidth, with all interactions operating concurrently in the analog domain. To reduce area and routing overhead, we are finalizing a banded sparse architecture for tape-out.
W2-b / Q3. Energy descent in non-ideal systems.
Strict monotone descent holds only for the ideal system. In non-ideal systems such as the hardware, the energy fluctuates but still descends, so energy minimization remains the core mechanism. Noise can even help: Ising machines for optimization inject noise on purpose to escape local minima. We show the energy trajectories of image generation from the same initial state in the revised paper (App. F): the ideal system descends smoothly, and the non-ideal system follows the same trend with small fluctuations. Saturation removes terms from the energy, so the energy stays non-increasing (Sec. 2.2). Models trained with asymmetric couplings keep their quality (clean-FID 5.96 and 5.39, App. B).
References
[1] H. Lee, J. P. Kim, and S. Kim. Fundamental challenges, physical implementations, and integration strategies for Ising machines in large-scale optimization tasks. Advanced Electronic Materials, 12(17):e00682, 2026.
[2] T. Zhang, Q. Tao, B. Liu, A. Grimaldi, E. Raimondo, M. Jiménez, M. J. Avedillo, J. Nuñez, B. Linares-Barranco, T. Serrano-Gotarredona, G. Finocchio, and J. Han. A review of Ising machines implemented in conventional and emerging technologies. IEEE Transactions on Nanotechnology, 23:704–717, 2024.
Detailed breakdown
| (C) Group (time) | Operation | Power (mW), Case A / Case B | Energy (nJ), Case A / Case B |
|---|---|---|---|
| Storage | Coupling coefficients, 10 MiB | 1.15 | 66.4 / 66.5 |
| Conditioning bias LUT, 25 KiB | 0.0173 | 1.0 | |
| DS results, 4 KiB | 0.00278 | 0.2 | |
| Seed retention | 0.344 | 19.8 / 19.9 | |
| Bias programming (10 ns) | Local class lookup and code update | 957 | 9.6 |
| Initialization (10 ns) | Shared random-bit source | 0.320 | 0.0032 |
| 4096-bit register shift | 6.31 | 0.0631 | |
| State loading and wire charging | 2,280 | 22.8 | |
| DS core (10.1 / 80.8 ns) | Spin evolution | 1,230 | 12.4 / 99.3 |
| Coupling-network evolution | 6,790 / 106 | 68.5 / 8.6 | |
| Conditioning bias | 34.4 | 0.3 / 2.8 | |
| DS interconnect (during DS core) | Layout parasitic effect | 136 / 17.0 | 1.4 |
| Readout (25 ns) | Sample / hold and voltage ADC | 4,510 | 112.6 |
| Readout source / buffer | 1,230 | 30.7 | |
| DS interface (320 ns) | Terminal SRAM access | 71.0 | 22.7 |
| Terminal data wires | 65.6 | 21.0 | |
| Decoder (57,300 ns) | Decoder computation | — | 5,740 |
| Control and communication | DS controller (during DS core) | 24.1 | 0.2 / 1.9 |
| Initialization / readout sequencer (365 ns) | 0.376 | 0.1 | |
| Seed-register clock distribution | 95.6 | 1.0 | |
| Seed-register interstage wires | 4.10 | 0.041 | |
| Class-select / reset / capture broadcast (40 ns) | 191 | 7.7 | |
| Total (57,675 / 57,746 ns) | — | 6,138.6 / 6,169.8 |
Thank you for the detailed response. It addresses most of my concerns, and I will raise my score from 2 to 3.
Thank you for your positive feedback, which have helped us further strengthen the paper. We sincerely appreciate your time and attention.
Response to Reviewer tVwc (Part 1/2)
Thank you for the detailed review. Your questions helped us make the hardware analysis more complete and auditable. We address each point below and will include the new results in the revised paper.
Energy and latency. We added an end-to-end breakdown with the same inclusion rules for our design and the GPU baseline. Tables A and B give coarse-grained energy and latency breakdowns, and Table C (at the end of the comment) gives a fine-grained breakdown by operation. Some highlights:
Hardware evaluation. We evaluate the complete DS implementation in Cadence Virtuoso AMS. The digital control logic is synthesized with Synopsys Design Compiler and placed and routed with IC Compiler. The resulting gate-level netlist is integrated into the mixed-signal simulation. Analog and digital parasitics are extracted using Mentor Calibre and Synopsys IC Compiler. SRAM access and retention power are estimated using the TSMC memory compiler. The mixed-signal circuit power is obtained from AMS simulation results.
GPU measurement. The timing and energy of all GPU baselines and decoders are measured on one NVIDIA L40S (350 W) with PyTorch 2.11, CUDA 12.8, FP32 weights, and TF32 tensor cores. We time each model from noise or labels to the output image, after warm-up. For our models (the -GPU family), only the decoder runs on the same L40S. Energy comes from the NVML board counter, with idle power included.