Pith. sign in

REVIEW 3 major objections 6 minor 44 references

TRACE: Learning to Compute on Circuit Graphs

T0 review · 3 major / 6 minor · reviewed 2026-08-15 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗ Review JSON

Pith's one-line read TRACE claims that modeling each computation step as an ordered prefix sequence and learning the function shift yields more accurate circuit behavior models across RTL, AIG, and netlists.

desk verdict Strong new circuit encoder with consistent wins across RTL/AIG/netlists; the function-shift inference loop has a real train/test mismatch that needs analysis before the numbers are taken at face value. read the letter →

arxiv 2509.21886 v3 pith:GUVTGCO5 submitted 2025-09-26 cs.AI

Ziyang Zheng , Jiaying Zhu , Jingyi Zhou , Qiang Xu This is my paper · ORCID
classification cs.AI
keywords circuitrepresentationlearninghierarchicaltransformerfunctionshiftlogic-1probabilitypredictionAnd-InvertergraphsRTLgraphretrievalpost-mappingnetlistcomputational
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

The paper tries to establish that the reason graph neural networks fail to model circuit behavior is architectural, not just a matter of training data. It claims that permutation-invariant message passing cannot represent ordered, operator-specific input interactions such as a multiplexer's select logic, and that flattened Transformers discard the nested dependency structure of computation. TRACE replaces both with a hierarchical Transformer that reads each computation step as an ordered prefix sequence and processes the graph level by level. For regression, it introduces function shift learning: the model predicts only the gap between the true global function and a cheap local approximation that treats inputs as independent, and reconstructs global values level by level at inference. Across RTL, AIG, and post-mapping netlist benchmarks, the paper reports consistent improvements over prior circuit-learning architectures, which matters because logic-1 probabilities, transition probabilities, and functional embeddings feed into power, timing, and verification tools.

What carries the argument

The load-bearing object is the per-step prefix sequence $v_i^k = [v_i^k, u_{j_1}, \dots, u_{j_{|N(v_i^k)|}}]$, with the operator token first and its ordered predecessors after it, processed by a Transformer encoder with positional encodings; this makes input order explicit and lets attention mix operator and inputs before any aggregation. The second mechanism is function shift learning: with the global function $y_{\text{global}}^\phi = \mathbb{E}_{x \sim D}[\phi(x_1,\dots,x_k)]$ and the local approximation $y_{\text{local}}^\phi = \phi(\mathbb{E}_{x\sim D}[x_1],\dots,\mathbb{E}_{x\sim D}[x_k])$, the training target is $y_{\text{FSL}}^\phi = y_{\text{global}}^\phi - y_{\text{local}}^\phi$, so the model learns only the reconvergence-induced correction. During inference the predicted shift is added to the local function built from previously estimated predecessor probabilities, level by level, in Algorithm 1. Together these replace permutation-invariant aggregation with position-aware, operator-specific computation and separate a node's intrinsic behavior from its context.

What would settle it

Take a combinational AIG with a deliberately long reconvergent chain, e.g., $c_k = x_1 \wedge x_2 \wedge \dots \wedge x_k$ built so each intermediate shares a variable with the next, simulate exhaustively to get exact logic-1 probabilities, and compare TRACE's level-by-level predictions; if MAE grows with chain length while single-shift accuracy stays flat, the inference recursion is compounding error.

Watch

Extended reading notes

Core claim

In the paper's own terms, computation in a circuit graph should be learned the way it is executed: level by level, each operator reading its output value from the embeddings of its predecessors. TRACE encodes every computation step as the ordered prefix sequence $[\text{operator}, \text{input}_1, \text{input}_2, \dots]$ and applies a Transformer encoder with positional encoding to that sequence, taking the first token's output as the updated node embedding. For prediction, it defines the global function as the expectation of the operator applied to the input distribution, and the local function as the operator applied to the expectations of the inputs; the model is trained to regress the difference $y_{\text{FSL}} = y_{\text{global}} - y_{\text{local}}$. At inference, Algorithm 1 reconstructs each node's global probability as the predicted shift plus the local function computed from already-estimated predecessor probabilities. The paper reports that this combination outperforms prior MPNN and Graph Transformer baselines across RTL graphs, combinational and sequential AIGs, and post-mapping netlists on both retrieval and regression tasks.

Load-bearing premise

The model assumes it can compute the function-shift training target from the available simulation data, and that its own level-by-level estimates of predecessor global probabilities stay accurate enough that errors do not compound when reconstructing deeper nodes.

Editorial extensions

If this is right

  • If TRACE's claim is right, circuit representation models built on permutation-invariant aggregation face a hard ceiling for position-aware operators, and the ordered-prefix encoder is the structural fix.
  • Function shift learning turns a hard global regression into a residual correction, so the reported ablation gains (e.g., MAE dropping from 0.036 to 0.013 on PM netlist logic-1 probability) are explained by the decoupling and should transfer to other node-level circuit statistics.
  • A single hierarchical Transformer with bounded per-node sequence lengths covers RTL, AIG, and netlist modalities without modality-specific graph rewiring, because the only structural assumption is operator-determined in-degree.
  • On sequential circuits, treating registers as pseudo primary inputs and removing feedback loops reduces cyclic behavior to level-by-level computation, preserving the reported gains on transition probability prediction.
  • The near-perfect retrieval scores reported (Rec@10 above 99% on RTL and AIG) imply the learned embeddings separate functionally distinct circuits almost perfectly, making them usable for functional clustering and similarity screening.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • Beyond the tested circuit settings, the ordered-prefix encoding should transfer to software control/data-flow graphs and bit-vector formulas in prefix form, which the paper mentions as a natural fit, but no experiments there are reported.
  • The paper leaves implicit that the function-shift training target depends on the input distribution used to compute the global function; a model trained under uniform Bernoulli inputs may need retraining or explicit conditioning before it generalizes to circuits with skewed signal probabilities.
  • A testable extension would measure error accumulation in Algorithm 1 by comparing each level's estimated global probabilities against exact simulation; if errors compound, an end-to-end residual or multi-pass refinement could be added without changing the backbone.
  • The contrastive results suggest the embedding space is nearly aligned with functional equivalence, which could support functional hashing or equivalence checking as downstream tasks, but the paper itself evaluates only retrieval and node-level regression.
Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

3 major / 6 minor

Summary. The paper introduces TRACE, a Transformer-based architecture for learning the functional behavior of circuit graphs. The backbone encodes each computation step as an ordered prefix sequence [operator, input 1, input 2, ...] and processes it with a small Transformer equipped with positional encodings, applied level by level along the graph's topological order. For node-level predictive tasks, the paper proposes Function Shift Learning (FSL), in which the model regresses the difference y_FSL = y_global - y_local, where y_local is the operator evaluated on the marginal probabilities of its predecessors and y_global is the true expectation of the operator's output. At inference, global estimates are reconstructed level by level by adding the predicted shift to a local term computed from previously estimated predecessor globals. The paper evaluates TRACE on RTL graphs, combinational and sequential AIGs, and post-mapping netlists, for both contrastive retrieval and predictive regression tasks, reporting consistent improvements over MPNN and graph-transformer baselines.

Significance. If the empirical claims hold, TRACE provides a genuinely different inductive bias for circuit representation learning: position-aware, operator-specific processing of a node's fan-in, plus a residual target that separates a mean-field local estimate from the correction caused by reconvergent dependencies. The benchmark coverage is broad, spanning three circuit modalities and both retrieval and regression tasks, and the FSL ablation in Table 4 supports the usefulness of the proposed objective. The paper does not ship code, proofs, or seed-level reproducibility, so the evidence is entirely empirical. Two aspects of the predictive pipeline are not yet documented: how ground-truth global functions are computed for large circuits, and how the iterative FSL reconstruction behaves when predecessor estimates are imperfect. Because these aspects directly govern the headline numbers, the current submission is not fully self-contained; the central idea is promising but needs additional analysis before the strength of the claims can be endorsed.

major comments (3)
  1. [Section 3.3, Algorithm 1] There is a training/inference mismatch in the FSL reconstruction. During training, Eq. (5) defines the target y_FSL_i = y_global_i - y_local_i with y_local_i evaluated at ground-truth predecessor global functions. During inference, Algorithm 1 line 6 computes the local term from the model's own estimates, y_hat_global_u1, ..., y_hat_global_uk. The predicted shift was trained to correct the exact local term, so when the local term is evaluated at estimated inputs the correction is miscalibrated, and errors can propagate through the level-by-level loop. This is load-bearing because the paper's central claim is that FSL 'decouples' global context from local computation; that decoupling is only valid if the local computation can be evaluated without dependence on the model's uncertain outputs. Please report per-level MAE/R2 for deep circuits (e.g., Table 6, combinational AIG max depth 2657), compare training-time and inference-time local inputs, and either modify the training procedure (e.g., scheduled sampling or noise injection) or demonstrate quantitatively that error compounding is negligible.
  2. [Section 3.3, Definition 1, Appendix E] The paper never specifies how the ground-truth global function y_global is computed. Definition 1 is an exact expectation over the input distribution, and the paper itself notes that direct computation requires O(2^k) enumeration. For the circuits in Table 6, with up to 45,409 nodes and depth 2,657, exact enumeration is not obviously feasible. If the labels come from random simulation, the number of input vectors, the simulation procedure, and whether y_local is computed from the same simulated marginals must be stated. Without this information, the FSL targets and the reported R2 and MAE values are not reproducible, and the metrics may be biased by label noise. Please document the label-generation procedure precisely.
  3. [Tables 1-4] All reported metrics are single numbers; Tables 1-4 contain no standard deviations, number of seeds, or details of evaluation splits. Some of the margins are small, for example Table 2 reports logic-1 R2 of 0.989 for TRACE versus 0.984 for DeepGate4, and Table 4 reports AIG similarity R2 of 0.533 versus 0.500 for TRACE without FSL. Without run-level variance, the claim that TRACE 'consistently and substantially outperforms' prior approaches is not statistically grounded for those close comparisons. Please report mean and standard deviation over at least three to five runs and specify how test circuits are split from training circuits.
minor comments (6)
  1. [Section 1, Figure 1] The claim that MPNNs are 'architecturally impossible' to model position-aware operators is too strong; standard MPNNs with appropriate node features and sufficient depth can represent order-dependent functions in principle. Consider softening to 'a single-layer permutation-invariant aggregation cannot distinguish input order' or providing a formal statement.
  2. [Equation (3)] The symbol v_i^k is used both for a node and for the sequence constructed from that node and its inputs; this is confusing. Please use distinct notation, e.g., s_i^k = [v_i^k, u_j1, ..., u_jm].
  3. [Algorithm 1 and Section 2.1] The local function phi_v is not defined for each operator type, particularly for word-level RTL operators such as Add, Equal, and Mux when applied to probabilities. Please state how phi_v is evaluated for each operator class, especially when signals are multi-bit.
  4. [Section 3.4] The 'functionally equivalent transformation' used to create positive contrastive samples is not specified. Please list the exact transformations applied for RTL, AIG, and PM netlist modalities, as this is essential for reproducing the retrieval experiments.
  5. [Appendix E.2] For functional similarity prediction, the ground truth is based on partial truth tables over a sampled set of input patterns, but the number of patterns and the number of node pairs are not reported. Please provide these values so the label quality can be assessed.
  6. [Tables 5 and 6] The 'Depth' statistic is not defined. Please clarify whether it is the maximum logic level from Eq. (2), and note that for sequential circuits the pseudo-PI treatment of registers affects the depth computation.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: FSL targets come from external simulation and the hierarchical transformer is an independent architectural contribution; self-citations are baselines and conventions.

full rationale

The paper's derivation chain is self-contained against external ground truth. The function shift target (y_FSL = y_global - y_local, Definitions 1-2, Eq. 5) is computed from externally simulated global probabilities and the closed-form local approximation, not from the model's own outputs, so predicting it is not predicting a fitted constant. Algorithm 1 reconstructs y_hat_global as y_hat_FSL + local function; this is a decomposition identity, not a circular definition, because y_hat_FSL is learned from data and the local function is evaluated on inputs (ground-truth at training, estimated at inference). The train/inference mismatch (estimated predecessor globals enter the local function) is a potential error-compounding robustness issue, but it is not a circularity: the paper never defines the target as the model's own output. The hierarchical transformer (Eq. 3-4) is an independent architectural proposal with positional encodings; no load-bearing premise is justified solely by self-citation. The many overlapping-author references (DeepGate*, DeepSeq2, ForgeEDA, DeepCell) are used as baselines, dataset sources, and standard preprocessing conventions (e.g., treating flip-flops as pseudo-PIs), not as the argument for TRACE's central claims. Therefore no step reduces by construction to its own inputs.

Assumptions & free parameters 0 free parameters · 4 assumptions · 1 invented entities

The central empirical comparison is not a parameter-free derivation; it relies on standard learned weights and several domain assumptions. No free constants are fitted to the data in closed form. The main unstated bets are the DAG reduction for sequential circuits and the accuracy of the signal probability labels used to train the function shift regression.

assumptions (4)
  • domain assumption Sequential circuits may be treated as DAGs by cutting feedback edges and treating registers as pseudo primary inputs.
    Invoked in Section 3.2 footnote 1; the logic level computation and the level-by-level inference in Algorithm 1 both depend on this reduction being functionally faithful.
  • domain assumption Exact global logic-1 probabilities for training nodes can be pre-computed or simulated accurately enough to serve as labels.
    Section 3.3 claims ground-truth global and local functions 'can be pre-computed from the training data'; the paper does not specify whether this is exhaustive enumeration, BDDs, or Monte Carlo simulation, which affects label noise.
  • ad hoc to paper The Transformer output at the first token position after attending to the operator and its ordered inputs is a sufficient representation of the computation result.
    Equation (4) uses Transformer(v_i^k + pos)[0]; this is a design choice, not a proven property.
  • domain assumption Positive samples in the contrastive task are created by transformations that preserve circuit function.
    Section 3.4 states this but does not list the transformations or verify equivalence for RTL, AIG, and netlist modalities.
invented entities (1)
  • Function shift y_FSL = y_global - y_local
    purpose: Training target that separates a node's input-independent approximation from the correlation-induced correction.
    It is an internal construct defined in Section 3.3; there is no independent measurement of the function shift outside the paper's own definitions and benchmarks.

how reviews work

0 comments
Cite this review

Pith. "Pith review of TRACE: Learning to Compute on Circuit Graphs." pith.science (2026). https://pith.science/paper/GUVTGCO5

@misc{pith2026250921886,
  author       = {Pith},
  title        = {Pith review of: TRACE: Learning to Compute on Circuit Graphs},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/GUVTGCO5}},
  note         = {Machine review of arXiv:2509.21886}
}
read the original abstract

Learning to compute, the ability to model the functional behavior of a circuit graph, is a fundamental challenge for graph representation learning. Yet, the dominant paradigm is architecturally mismatched for this task. This flawed assumption, central to mainstream message passing neural networks (MPNNs) and their conventional Transformer-based counterparts, prevents models from capturing the position-aware, hierarchical nature of computation. To resolve this, we introduce TRACE, a new paradigm built on an architecturally sound backbone and a principled learning objective. First, TRACE employs a Hierarchical Transformer that mirrors the step-by-step flow of computation, providing a faithful architectural backbone that replaces the flawed permutation-invariant aggregation. Second, we introduce function shift learning, a novel objective that decouples the learning problem. Instead of predicting the complex global function directly, our model is trained to predict only the function shift, the discrepancy between the true global function and a simple local approximation that assumes input independence. We validate this paradigm on various circuits modalities, including Register Transfer Level graphs, And-Inverter Graphs and post-mapping netlists. Across a comprehensive suite of benchmarks, TRACE substantially outperforms all prior architectures. These results demonstrate that our architecturally-aligned backbone and decoupled learning objective form a more robust paradigm for the fundamental challenge of learning the functional behavior of a circuit graph.

Figures

Figures reproduced from arXiv: 2509.21886 by the authors.

Figure 1
Figure 1. The architectural failure of MPNNs on computational graphs. [PITH_FULL_IMAGE:figures/full_fig_p002_1.png] view at source ↗
Figure 2
Figure 2. Illustration of a Computational Graph and Its Computation Process. This figure demon [PITH_FULL_IMAGE:figures/full_fig_p003_2.png] view at source ↗
Figure 3
Figure 3. Overview of our proposed framework. Left: A circuit graph, represented in both a graph view and its equivalent prefix notation, is encoded by a Hierarchical Transformer to model the computation process. Right: For predictive tasks, we introduce Function Shift Learning (FSL). Instead of directly regressing the global function, the model captures the difference between the global and local functions: y F SL = y global… view at source ↗
Figures from the paper (1 more)
Figure 4
Figure 4. Figure 4: In-degree distribution. The x-axis represents the node in-degree and the y-axis represents [PITH_FULL_IMAGE:figures/full_fig_p014_4.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

44 extracted references · 19 canonical work pages

  1. [1]

    Iwls 2005 benchmarks

    Christoph Albrecht. Iwls 2005 benchmarks. In International Workshop for Logic Synthesis (IWLS), volume 9, 2005

    work page 2005

  2. [2]

    On the bottleneck of graph neural networks and its practical implications

    Uri Alon and Eran Yahav. On the bottleneck of graph neural networks and its practical implications. arXiv preprint arXiv:2006.05205, 2020

    arXiv 2006

  3. [3]

    Btor2 , btormc and boolector3.0

    ArminBiere, AinaNiemetz, MathiasPreiner, and CliffordWolf. Btor2 , btormc and boolector3.0. Springer, Cham, 2018

    work page 2018

  4. [4]

    Notes on the iscas'89 benchmark circuits

    Franc Brglez, David Bryan, and Krzysztof Kozminski. Notes on the iscas'89 benchmark circuits. Technical report, Technical report, MCNC, 1989. Online http://www. cbl. ncsu. edu/CBL Docs …, 1989

    work page 1989

  5. [5]

    Large circuit models: opportunities and challenges

    Lei Chen, Yiqi Chen, Zhufei Chu, Wenji Fang, Tsung-Yi Ho, Ru Huang, Yu Huang, Sadaf Khan, Min Li, Xingquan Li, et al. Large circuit models: opportunities and challenges. Science China Information Sciences, 67 0 (10): 0 200402, 2024

    work page 2024

  6. [6]

    Rt-level itc'99 benchmarks and first atpg results

    Fulvio Corno, Matteo Sonza Reorda, and Giovanni Squillero. Rt-level itc'99 benchmarks and first atpg results. IEEE Design & Test of computers, 17 0 (3): 0 44--53, 2002

    work page 2002

  7. [7]

    Less is more: Hop-wise graph attention for scalable and generalizable learning on circuits

    Chenhui Deng, Zichao Yue, Cunxi Yu, Gokce Sarar, Ryan Carey, Rajeev Jain, and Zhiru Zhang. Less is more: Hop-wise graph attention for scalable and generalizable learning on circuits. In Proceedings of the 61st ACM/IEEE Design Automation Conference, pp.\ 1--6, 2024

    work page 2024

  8. [8]

    NetTAG: A Multimodal RTL-and-Layout-Aligned Netlist Foundation Model via Text-Attributed Graph

    Wenji Fang, Wenkai Li, Shang Liu, Yao Lu, Hongce Zhang, and Zhiyao Xie. Nettag: A multimodal rtl-and-layout-aligned netlist foundation model via text-attributed graph. arXiv preprint arXiv:2504.09260, 2025 a

    work page Pith review arXiv 2025

Show all 44 references
  1. [9]

    Circuitfusion: multimodal circuit representation learning for agile chip design

    Wenji Fang, Shang Liu, Jing Wang, and Zhiyao Xie. Circuitfusion: multimodal circuit representation learning for agile chip design. arXiv preprint arXiv:2505.02168, 2025 b

    2025 arXiv

  2. [10]

    A self-supervised, pre-trained, and cross-stage-aligned circuit encoder provides a foundation for various design tasks

    Wenji Fang, Shang Liu, Hongce Zhang, and Zhiyao Xie. A self-supervised, pre-trained, and cross-stage-aligned circuit encoder provides a foundation for various design tasks. In Proceedings of the 30th Asia and South Pacific Design Automation Conference, pp.\ 505--512, 2025 c

    2025

  3. [11]

    Neural message passing for quantum chemistry

    Justin Gilmer, Samuel S Schoenholz, Patrick F Riley, Oriol Vinyals, and George E Dahl. Neural message passing for quantum chemistry. In International conference on machine learning, pp.\ 1263--1272. Pmlr, 2017

    2017

  4. [12]

    Inductive representation learning on large graphs

    Will Hamilton, Zhitao Ying, and Jure Leskovec. Inductive representation learning on large graphs. Advances in neural information processing systems, 30, 2017

    2017

  5. [13]

    Deepseq2: Enhanced sequential circuit learning with disentangled representations

    Sadaf Khan, Zhengyuan Shi, Ziyang Zheng, Min Li, and Qiang Xu. Deepseq2: Enhanced sequential circuit learning with disentangled representations. In Proceedings of the 30th Asia and South Pacific Design Automation Conference, pp.\ 498--504, 2025

    2025

  6. [14]

    Semi-supervised classification with graph convolutional networks

    TN Kipf. Semi-supervised classification with graph convolutional networks. arXiv preprint arXiv:1609.02907, 2016

    2016 arXiv

  7. [15]

    Deepgate: Learning neural representations of logic gates

    Min Li, Sadaf Khan, Zhengyuan Shi, Naixing Wang, Huang Yu, and Qiang Xu. Deepgate: Learning neural representations of logic gates. In Proceedings of the 59th ACM/IEEE Design Automation Conference, pp.\ 667--672, 2022

    2022

  8. [17]

    Deeper insights into graph convolutional networks for semi-supervised learning

    Qimai Li, Zhichao Han, and Xiao-Ming Wu. Deeper insights into graph convolutional networks for semi-supervised learning. In AAAI, volume 32, 2018

    2018

  9. [18]

    Deepcircuitx: A comprehensive repository-level dataset for rtl code understanding, generation, and ppa analysis

    Zeju Li, Changran Xu, Zhengyuan Shi, Zedong Peng, Yi Liu, Yunhao Zhou, Lingfeng Zhou, Chengyu Ma, Jianyuan Zhong, Xi Wang, et al. Deepcircuitx: A comprehensive repository-level dataset for rtl code understanding, generation, and ppa analysis. arXiv preprint arXiv:2502.18297, 2025

    2025 arXiv

  10. [19]

    Polargate: Breaking the functionality representation bottleneck of and-inverter graph neural network

    Jiawei Liu, Jianwang Zhai, Mingyu Zhao, Zhe Lin, Bei Yu, and Chuan Shi. Polargate: Breaking the functionality representation bottleneck of and-inverter graph neural network. In 2024 IEEE/ACM International Conference on Computer-Aided Design (ICCAD), 2024

    2024

  11. [20]

    Ithemal: Accurate, portable and fast basic block throughput estimation using deep neural networks

    Charith Mendis, Alex Renda, Saman Amarasinghe, and Michael Carbin. Ithemal: Accurate, portable and fast basic block throughput estimation using deep neural networks. In 36th International Conference on Machine Learning, ICML 2019, 36th International Conference on Machine Learn...

    2019

  12. [21]

    Songhori, Shen Wang, Young-Joon Lee, Eric Johnson, Omkar Pathak, Azade Nazi, Jiwoo Pak, Andy Tong, Kavya Srinivasa, Will Hang, Emre Tuncer, Quoc V

    Azalia Mirhoseini, Anna Goldie, Mustafa Yazgan, Joe Wenjie Jiang, Ebrahim M. Songhori, Shen Wang, Young-Joon Lee, Eric Johnson, Omkar Pathak, Azade Nazi, Jiwoo Pak, Andy Tong, Kavya Srinivasa, Will Hang, Emre Tuncer, Quoc V. Le, James Laudon, Richard Ho, Roger Carpenter, and J...

    2021

  13. [22]

    Representation learning with contrastive predictive coding

    Aaron van den Oord, Yazhe Li, and Oriol Vinyals. Representation learning with contrastive predictive coding. arXiv preprint arXiv:1807.03748, 2018

    2018 arXiv

  14. [23]

    Recipe for a general, powerful, scalable graph transformer

    Ladislav Ramp \'a s ek, Michael Galkin, Vijay Prakash Dwivedi, Anh Tuan Luu, Guy Wolf, and Dominique Beaini. Recipe for a general, powerful, scalable graph transformer. Advances in Neural Information Processing Systems, 35: 0 14501--14515, 2022

    2022

  15. [24]

    Daniel Selsam, Matthew Lamm, Benedikt B \"u nz, Percy Liang, Leonardo Mendonça de Moura, and David L. Dill. Learning a sat solver from single-bit supervision. ArXiv, abs/1802.03685, 2018. URL https://api.semanticscholar.org/CorpusID:3632319

    2018 arXiv

  16. [25]

    Deepgate2: Functionality-aware circuit representation learning

    Zhengyuan Shi, Hongyang Pan, Sadaf Khan, Min Li, Yi Liu, Junhua Huang, Hui-Ling Zhen, Mingxuan Yuan, Zhufei Chu, and Qiang Xu. Deepgate2: Functionality-aware circuit representation learning. In 2023 IEEE/ACM International Conference on Computer Aided Design (ICCAD), pp.\ 1--9....

    2023

  17. [26]

    Deepgate3: Towards scalable circuit representation learning

    Zhengyuan Shi, Ziyang Zheng, Sadaf Khan, Jianyuan Zhong, Min Li, and Qiang Xu. Deepgate3: Towards scalable circuit representation learning. arXiv preprint arXiv:2407.11095, 2024

    2024 arXiv

  18. [27]

    Forgeeda: A comprehensive multimodal dataset for advancing eda

    Zhengyuan Shi, Zeju Li, Chengyu Ma, Yunhao Zhou, Ziyang Zheng, Jiawei Liu, Hongyang Pan, Lingfeng Zhou, Kezhi Li, Jiaying Zhu, et al. Forgeeda: A comprehensive multimodal dataset for advancing eda. arXiv preprint arXiv:2505.02016, 2025 a

    2025 arXiv

  19. [28]

    Deepcell: Multiview representation learning for post-mapping netlists

    Zhengyuan Shi, Chengyu Ma, Ziyang Zheng, Lingfeng Zhou, Hongyang Pan, Wentao Jiang, Fan Yang, Xiaoyan Yang, Zhufei Chu, and Qiang Xu. Deepcell: Multiview representation learning for post-mapping netlists. arXiv preprint arXiv:2502.06816, 2025 b

    2025 arXiv

  20. [29]

    Graph attention networks

    Petar Veli c kovi \'c , Guillem Cucurull, Arantxa Casanova, Adriana Romero, Pietro Lio, and Yoshua Bengio. Graph attention networks. arXiv preprint arXiv:1710.10903, 2017

    2017 arXiv

  21. [30]

    Fgnn2: A powerful pre-training framework for learning the logic functionality of circuits

    Ziyi Wang, Chen Bai, Zhuolun He, Guangliang Zhang, Qiang Xu, Tsung-Yi Ho, Yu Huang, and Bei Yu. Fgnn2: A powerful pre-training framework for learning the logic functionality of circuits. IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems, 2024

    2024

  22. [31]

    Circuit representation learning with masked gate modeling and verilog-aig alignment

    Haoyuan Wu, Haisheng Zheng, Yuan Pu, and Bei Yu. Circuit representation learning with masked gate modeling and verilog-aig alignment. arXiv preprint arXiv:2502.12732, 2025

    2025 arXiv

  23. [32]

    Gamora: Graph learning based symbolic reasoning for large-scale boolean networks

    Nan Wu, Yingjie Li, Cong Hao, Steve Dai, Cunxi Yu, and Yuan Xie. Gamora: Graph learning based symbolic reasoning for large-scale boolean networks. In 2023 60th ACM/IEEE Design Automation Conference (DAC), pp.\ 1--6. IEEE, 2023 a

    2023

  24. [33]

    Difformer: Scalable (graph) transformers induced by energy constrained diffusion

    Qitian Wu, Chenxiao Yang, Wentao Zhao, Yixuan He, David Wipf, and Junchi Yan. Difformer: Scalable (graph) transformers induced by energy constrained diffusion. arXiv preprint arXiv:2301.09474, 2023 b

    2023 arXiv

  25. [34]

    Sgformer: Simplifying and empowering transformers for large-graph representations

    Qitian Wu, Wentao Zhao, Chenxiao Yang, Hengrui Zhang, Fan Nie, Haitian Jiang, Yatao Bian, and Junchi Yan. Sgformer: Simplifying and empowering transformers for large-graph representations. Advances in Neural Information Processing Systems, 36: 0 64753--64773, 2023 c

    2023

  26. [35]

    Preplacement net length and timing estimation by customized graph neural network

    Zhiyao Xie, Rongjian Liang, Xiaoqing Xu, Jiang Hu, Chen-Chia Chang, Jingyu Pan, and Yiran Chen. Preplacement net length and timing estimation by customized graph neural network. IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems, 41 0 (11): 0 4667--4...

    2022

  27. [36]

    How powerful are graph neural networks? arXiv preprint arXiv:1810.00826, 2018

    Keyulu Xu, Weihua Hu, Jure Leskovec, and Stefanie Jegelka. How powerful are graph neural networks? arXiv preprint arXiv:1810.00826, 2018

    2018 arXiv

  28. [37]

    Revisiting semi-supervised learning with graph embeddings

    Zhilin Yang, William Cohen, and Ruslan Salakhudinov. Revisiting semi-supervised learning with graph embeddings. In International conference on machine learning, pp.\ 40--48. PMLR, 2016

    2016

  29. [38]

    Jiang, Luca Amarú, Alan Mishchenko, and Robert Brayton

    He-Teng Zhang, Jie-Hong R. Jiang, Luca Amarú, Alan Mishchenko, and Robert Brayton. Deep integration of circuit simulator and sat solver. In 2021 58th ACM/IEEE Design Automation Conference (DAC), pp.\ 877--882, 2021. doi:10.1109/DAC18074.2021.9586331

    2021

  30. [39]

    Grannite: Graph neural network inference for transferable power estimation

    Yanqing Zhang, Haoxing Ren, and Brucek Khailany. Grannite: Graph neural network inference for transferable power estimation. In 2020 57th ACM/IEEE Design Automation Conference (DAC), pp.\ 1--6, 2020. doi:10.1109/DAC18072.2020.9218643

    2020

  31. [40]

    Deepgate4: Efficient and effective representation learning for circuit design at scale

    Ziyang Zheng, Shan Huang, Jianyuan Zhong, Zhengyuan Shi, Guohao Dai, Ningyi Xu, and Qiang Xu. Deepgate4: Efficient and effective representation learning for circuit design at scale. arXiv preprint arXiv:2502.01681, 2025

    2025 arXiv

  32. [41]

    Rl-mul: Multiplier design optimization with deep reinforcement learning

    Dongsheng Zuo, Yikang Ouyang, and Yuzhe Ma. Rl-mul: Multiplier design optimization with deep reinforcement learning. In 2023 60th ACM/IEEE Design Automation Conference (DAC), pp.\ 1--6, 2023. doi:10.1109/DAC56929.2023.10247941

    2023

  33. [42]

    write newline

    " write newline "" before.all 'output.state := FUNCTION n.dashify 't := "" t empty not t #1 #1 substring "-" = t #1 #2 substring "--" = not "--" * t #2 global.max substring 't := t #1 #1 substring "-" = "-" * t #2 global.max substring 't := while if t #1 #1 substring * t #2 gl...

  34. [43]

    @esa (Ref

    \@ifxundefined[1] #1\@undefined \@firstoftwo \@secondoftwo \@ifnum[1] #1 \@firstoftwo \@secondoftwo \@ifx[1] #1 \@firstoftwo \@secondoftwo [2] @ #1 \@temptokena #2 #1 @ \@temptokena \@ifclassloaded agu2001 natbib The agu2001 class already includes natbib coding, so you should ...

  35. [44]

    \@lbibitem[] @bibitem@first@sw\@secondoftwo \@lbibitem[#1]#2 \@extra@b@citeb \@ifundefined br@#2\@extra@b@citeb \@namedef br@#2 \@nameuse br@#2\@extra@b@citeb \@ifundefined b@#2\@extra@b@citeb @num @parse #2 @tmp #1 NAT@b@open@#2 NAT@b@shut@#2 \@ifnum @merge>\@ne @bibitem@firs...

  36. [45]

    Yet, the dominant paradigm is architecturally mismatched for this task

    @open @close @open @close and [1] URL: #1 \@ifundefined chapter * \@mkboth \@ifxundefined @sectionbib * \@mkboth * \@mkboth\@gobbletwo \@ifclassloaded amsart * \@ifclassloaded amsbook * \@ifxundefined @heading @heading NAT@ctr thebibliography [1] @ \@biblabel @NAT@ctr \@bibset...

    1987

Pith tools

Reviewed August 15, 2026 · model on record in the stance chip above.