Pith. sign in

REVIEW 3 major objections 5 minor 51 references

RTL-Sequencer: Towards Scalable RTL Timing Prediction with the Sequence-based Paradigm

T0 review · 3 major / 5 minor · reviewed 2026-08-01 · deepseek-v4-flash

Pith's one-line read RTL timing prediction improves when logic cones are read as linear sequences instead of graphs.

desk verdict A sensible sequence-based alternative to graph timing models with promising results, but the claimed 'consistent' superiority lacks variance reporting and public artifacts. read the letter →

arxiv 2607.15830 v1 pith:SZ3Q6HUY submitted 2026-07-17 cs.AR cs.AIcs.LG

classification cs.ARcs.AIcs.LG
keywords RTLtimingpredictionlogicconesequencemodelbreadth-firsttraversallinearcomplexitystatespaceanalysisdesignautomation
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

This paper argues that RTL timing prediction should be treated as a sequence-learning problem rather than a graph-learning problem. It shows that a logic cone – the directed circuit subgraph feeding an output register – can be flattened by breadth-first traversal into a node sequence, and that modern linear sequence models trained on such sequences predict arrival time, worst negative slack, and total negative slack more accurately than graph neural networks or graph transformers. On 21 open-source RTL designs with 5-fold cross-validation, it reports AT MAPE 17.24%, WNS MAPE 17.66%, and TNS MAPE 23.92%, all better than the baselines, while maintaining linear computational complexity and a global receptive field. If this holds, designers could get accurate timing feedback before synthesis and layout, shortening the design loop.

What carries the argument

The key object is the BFT-linearized logic cone sequence. Each cone is traversed backward from the endpoint register, visiting nodes level by level; the resulting sequence carries a directional ordering that mirrors signal propagation. A linear sequence model (Mamba-2 in the implementation) updates a hidden state per node in linear time, giving each node implicit access to all predecessors. The four customizations – shuffling within BFT levels, bidirectional passes, differential depth-subtraction with a learned gate, and a GNN front-end – compensate for the topological information lost when a DAG is flattened into a fixed order.

What would settle it

If, on a synthetic logic cone where two nodes at the same BFT depth feed paths with very different delays, the predictor cannot separate their contributions – MAPE worsens as such same-depth collisions are introduced – the claimed information-preservation would fail. Concretely: generate cones with controlled same-depth sibling structure and compare RTL-Sequencer against a depth-first linearization baseline; if BFT no longer helps, the central mechanism is suspect.

Watch

Extended reading notes

Core claim

The central claim is that flattening each logic cone into an ordered sequence by inverse breadth-first traversal, then applying a linear sequence model, recovers enough of the cone's directed timing structure to outperform graph-based predictors. To make this work, the authors add four mechanisms: random shuffling of nodes within each BFT level as data augmentation, bidirectional sequence passes so later nodes influence earlier ones, a differentiable 'global minus depth-local' embedding subtraction that suppresses same-depth neighbors' spurious influence, and a hybrid pipeline that first runs a GNN for local structure then the sequence model for long-range depth. The result is a timing predi

Load-bearing premise

The whole approach rests on the assumption that flattening a logic cone into a breadth-first node sequence preserves enough of its directed timing-critical structure for a sequence model to recover it, even though BFT places unrelated same-depth nodes side by side.

Editorial extensions

If this is right

  • Early RTL timing feedback could become accurate enough to guide design changes before synthesis, shortening design iterations.
  • Timing prediction no longer needs to trade off receptive field against computational cost: the sequence model provides both global reach and linear scaling.
  • Training on very large cones – tens of thousands of nodes – becomes practical on modest hardware because memory and time grow linearly with cone size.
  • The hybrid graph–sequence design indicates that local structure and global depth are complementary signals, and both can be captured end-to-end.

Reading between the lines

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

  • The same BFT-linearization approach could extend to other cone-level prediction tasks in chip design, such as power or routability, where directed signal flow also matters and graph models face the same receptive-field limits.
  • The shuffling result suggests the model learns timing structure rather than a fixed node order; a stronger generalization test would measure performance on cones whose depth distribution lies far outside the training set.
  • One could probe the differential module by constructing synthetic cones where two same-depth nodes feed paths with very different delays; if the model still separates them, the depth-subtraction is doing the intended work.
Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

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

Referee Report

3 major / 5 minor

Summary. The paper proposes RTL-Sequencer, a sequence-based framework for RTL timing prediction. Logic cones are linearized by inverse breadth-first traversal and processed with Mamba-2 linear sequence models, augmented by four techniques: stochastic shuffling, bidirectional sequence modeling, a differential (global-minus-depth-local) operation, and a hybrid GNN-sequence architecture. Experiments on 21 open-source RTL designs with 5-fold cross-validation report AT MAPE 17.24%, WNS MAPE 17.66%, and TNS MAPE 23.92%, outperforming the evaluated graph-based and transformer baselines. The authors argue this establishes a scalable sequence-based paradigm with global receptive fields and linear complexity.

Significance. If the reported gains are reproducible and general, the paper makes a useful contribution: it challenges the graph-centric default in RTL timing prediction and shows that modern linear sequence models can be competitive or better with better scalability. The architecture is clearly described, the ablation study isolates each proposed component, and the comparison includes several strong recent baselines. The design-partitioned 5-fold protocol is appropriate for generalization claims. However, the central empirical claim is currently under-supported because no variance or significance information is reported, and the theoretical 'lossless encoding' statement in Section 3 is incorrect as written. The lack of released code/data also prevents independent verification of the baseline reimplementations.

major comments (3)
  1. [Sec. 5.2, Table 3] The central claim 'consistently surpasses prior approaches' rests on aggregate MAPE/R/R² values from 5-fold cross-validation over only 21 designs, so each test fold contains roughly 4 designs. No per-design or per-fold results, standard deviations across seeds, confidence intervals, or significance tests are reported. The AT MAPE improvement over CircuitFusion is 17.24% vs. 20.29%; without variance estimates this margin could lie within run-to-run variability or be dominated by one or two large designs. Please report per-design and per-fold errors, seed variance, and a paired significance test (e.g., Wilcoxon or bootstrap) on design-level errors. The same issue applies to Figure 4's claim that all sequence models surpass graph-based solutions. This is load-bearing because the paper's contribution is empirical.
  2. [Sec. 4.1, Sec. 4.4] The breadth-first linearization is the core enabler of the method, but the paper provides neither a formal guarantee nor a quantitative analysis that the linearization preserves enough directed timing information. Section 4.1 asserts BFT 'preserves the hierarchical progression of signal propagation,' while Section 4.4 concedes that BFT 'frequently places unrelated nodes at the same BFT depth,' making them adjacent in the sequence. Since a fixed BFT sequence with intra-level shuffling does not encode the DAG exactly, the model must recover timing-critical paths from an approximate linearization. The hybrid GNN module mitigates this, but the paper should evaluate sensitivity to the linearization (e.g., alternative traversals, different shuffling policies, or random permutations at inference) or provide an information-theoretic bound for the cone sizes considered. Without this, the general
  3. [Sec. 3, Eq. (2)-(3)] The statement that historical context is 'losslessly encoded by preceding nodes while maintaining linear computational complexity' is false for finite-dimensional hidden states: any fixed-size state is a lossy compression of an arbitrarily long prefix. This is not merely a wording issue, because the paper uses the lossless claim to motivate the entire sequence-based paradigm. Please replace it with a bounded-error or approximation statement, or show that for the finite logic-cone setting the state dimension is sufficient to represent the relevant timing information. The empirical results may still hold, but the theoretical motivation must be corrected.
minor comments (5)
  1. [Sec. 2] The definition of E is garbled: 'E denotes the set of nodes corresponding to registers and combinational logic elements' repeats the definition of V; E should be the set of edges.
  2. [Table 4 vs. Table 3] The TNS row for RTL-Sequencer reports R²=0.76 in Table 4 but R²=0.77 in Table 3. Please reconcile.
  3. [Sec. 4.4 and title] The mechanism is 'differential' (global minus depth-local), not merely 'differentiable.' The paper uses both terms interchangeably; recommend renaming to 'Differential Sequence Modeling' to match the cited Differential Transformer and to avoid confusion with differentiability.
  4. [References [3] and [4]] References [3] and [4] appear to be the same TF-Predictor paper listed twice with different years and venues. Please merge or disambiguate.
  5. [Sec. 4.2] The shuffling rule is ambiguous: 'randomized within each BFT depth level with 50% probability' could mean per-cone, per-level, or per-node. Please specify the exact stochastic procedure and how permutations are synchronized across child nodes.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the paper is an empirical method evaluation whose claims are tested against external STA ground truth and external baselines.

full rationale

No load-bearing step reduces to its inputs by construction. The prediction target (AT/WNS/TNS) is external ground truth from sign-off STA (Eq. 1), and the model is trained and evaluated via 5-fold cross-validation with strictly design-disjoint splits (Sec. 5.1), so no fitted parameter is renamed as a prediction. The BFT linearization and the four customization techniques (Secs. 4.1-4.5) are modeling assumptions and architectural designs, not derivations that presuppose the claimed performance improvement; even if BFT loses topological information, that is a correctness risk, not circularity. The paper contains self-citations (e.g., RTL-Timer [9], CircuitFusion [8], MasterRTL [11]), but these are used as prior benchmarks, dataset provenance, and related work, not as a uniqueness theorem or as an unverified premise that forces the proposed architecture. The absence of error bars and statistical significance tests is a reproducibility concern, not evidence of circularity. The central empirical claim stands or falls on the reported comparisons and ablations, which are independent of the paper's own outputs.

Assumptions & free parameters 5 free parameters · 5 assumptions · 0 invented entities

The central claim rests on empirical evaluation rather than a derivation, so the ledger is mostly about assumptions in data and modeling: BFT linearization preserves timing-critical information, sign-off STA labels are trustworthy, and the 21-design benchmark generalizes. Hyperparameters are hand-chosen without sensitivity analysis. No new physical or external entities are postulated; the architectural modules are not invented entities in the graviton sense.

free parameters (5)
  • hidden channel dimension = 32
    Chosen by hand in Sec. 5.1; no sensitivity analysis is given.
  • number of sequence blocks L = 3
    Chosen by hand in Sec. 5.1.
  • shuffling probability = 50%
    Applied to each training sample, Sec. 5.1.
  • learning rate = 1e-3
    Adam optimizer setting in Sec. 5.1.
  • per-GPU batch size = 8
    Training configuration in Sec. 5.1.
assumptions (5)
  • domain assumption BFT linearization preserves enough timing-critical signal-propagation information for sequence models to recover arrival times.
    Central modeling premise in Sec. 4.1; no formal guarantee is provided, and Sec. 4.4 acknowledges BFT places unrelated same-depth nodes adjacently.
  • domain assumption Ground-truth timing labels from sign-off STA (Synopsys PrimeTime after place-and-route) are accurate enough to supervise the model.
    Used to form the regression objective in Eq. (1) and the dataset in Sec. 5.1.
  • domain assumption The 21 open-source RTL designs with 5-fold design-split cross-validation are representative enough to support general claims about RTL timing prediction.
    The generalization claim rests on this benchmark, Sec. 5.1.
  • domain assumption Excluding post-synthesis features from all baselines is a fair apples-to-apples comparison.
    Sec. 5.1 says baselines are 'deliberately excluding auxiliary features derived from post-synthesis stages'; this may handicap methods designed to use such features.
  • domain assumption Random shuffling within BFT levels, with synchronized permutations across child nodes, preserves logical dependencies and timing closure.
    This is asserted in Sec. 4.2 as data augmentation; no proof or empirical check of logic preservation is given.

how reviews work

0 comments
Cite this review

Pith. "Pith review of RTL-Sequencer: Towards Scalable RTL Timing Prediction with the Sequence-based Paradigm." pith.science (2026). https://pith.science/paper/SZ3Q6HUY

@misc{pith2026260715830,
  author       = {Pith},
  title        = {Pith review of: RTL-Sequencer: Towards Scalable RTL Timing Prediction with the Sequence-based Paradigm},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/SZ3Q6HUY}},
  note         = {Machine review of arXiv:2607.15830}
}
read the original abstract

Accurate timing prediction at the register-transfer level (RTL) is a longstanding challenge in design automation. Existing graph-based methods struggle with limited receptive fields, high complexity, and a lack of signal directionality. We present RTL-Sequencer, a novel sequence-based paradigm that enables scalable RTL timing prediction via linearizing logic cones by breadth-first traversal and applying modern linear sequence models. Furthermore, sequence models are customized by four synergistic techniques, including sequence shuffling, bidirectional modeling, differentiable modeling, and a hybrid graph-sequence architecture. Extensive experiments demonstrate significant improvements of RTL-Sequencer over state-of-the-art baselines, advancing early-stage timing optimization.

Figures

Figures reproduced from arXiv: 2607.15830 by the authors.

Figure 1
Figure 1. RTL-Sequencer vs. conventional graph-based ap [PITH_FULL_IMAGE:figures/full_fig_p001_1.png] view at source ↗
Figure 2
Figure 2. Overview of RTL-Sequencer. (a) RTL data preprocessing includes extracting ground-truth timing metrics via commercial EDA tools and transforming RTL codes into BOGs. Subsequently, logic cones are extracted from the BOG. (b) Overall architecture of the proposed RTL-Sequencer framework, where the extracted logic cone from (a) is fed into a sequence modeling pipeline to predict timing metrics and facilitate neural netwo… view at source ↗
Figure 3
Figure 3. Scalability comparison among GCN, GT, GNN [PITH_FULL_IMAGE:figures/full_fig_p006_3.png] view at source ↗

Discussion (0). Sign in to comment.

Reference graph

Works this paper leans on

51 extracted references · 10 linked inside Pith

  1. [3]

    Peng Cao, Guoqing He, and Tai Yang. 2022. Tf-predictor: Transformer-based prerouting path delay prediction framework.IEEE Transactions on Computer- Aided Design of Integrated Circuits and Systems42, 7 (2022)

  2. [4]

    Peng Cao, Guoqing He, and Tai Yang. 2023. TF-Predictor: Transformer-Based Prerouting Path Delay Prediction Framework.IEEE Transactions on Computer- Aided Design of Integrated Circuits and Systems42, 7 (2023)

  3. [1]

    Erick Carvajal Barboza, Nishchal Shukla, Yiran Chen, and Jiang Hu. 2019. Ma- chine Learning-Based Pre-Routing Timing Prediction with Reduced Pessimism. In2019 56th ACM/IEEE Design Automation Conference (DAC)

  4. [2]

    Leo Breiman. 2001. Random forests.Machine learning45, 1 (2001)

  5. [5]

    Tianqi Chen and Carlos Guestrin. 2016. Xgboost: A scalable tree boosting system. InProceedings of the 22nd acm sigkdd international conference on knowledge discovery and data mining

  6. [6]

    Chhabria, Wenjing Jiang, Andrew B

    Vidya A. Chhabria, Wenjing Jiang, Andrew B. Kahng, and Sachin S. Sapatnekar

  7. [7]

    Tri Dao and Albert Gu. 2024. Transformers are ssms: Generalized models and efficient algorithms through structured state space duality.arXiv preprint arXiv:2405.21060(2024)

  8. [8]

    Wenji Fang, Shang Liu, Jing Wang, and Zhiyao Xie. 2025. Circuitfusion: mul- timodal circuit representation learning for agile chip design.arXiv preprint arXiv:2505.02168(2025)

Show all 51 references
  1. [9]

    Wenji Fang, Shang Liu, Hongce Zhang, and Zhiyao Xie. 2024. Annotating slack directly on your verilog: Fine-grained rtl timing evaluation for early optimization. InProceedings of the 61st ACM/IEEE Design Automation Conference

  2. [10]

    Wenji Fang, Shang Liu, Hongce Zhang, and Zhiyao Xie. 2025. A Self-Supervised, Pre-Trained, and Cross-Stage-Aligned Circuit Encoder Provides a Foundation for Various Design Tasks. InProceedings of the 30th Asia and South Pacific Design Automation Conference (ASPDAC ’25)

  3. [11]

    Wenji Fang, Yao Lu, Shang Liu, Qijun Zhang, Ceyu Xu, Lisa Wu Wills, Hongce Zhang, and Zhiyao Xie. 2023. Masterrtl: A pre-synthesis ppa estimation frame- work for any rtl design. In2023 IEEE/ACM International Conference on Computer Aided Design (ICCAD)

  4. [12]

    Albert Gu and Tri Dao. 2024. Mamba: Linear-time sequence modeling with selective state spaces. InFirst conference on language modeling

  5. [13]

    Zizheng Guo, Mingjie Liu, Jiaqi Gu, Shuhan Zhang, David Z Pan, and Yibo Lin

  6. [14]

    Pan, and Yibo Lin

    Zizheng Guo, Mingjie Liu, Jiaqi Gu, Shuhan Zhang, David Z. Pan, and Yibo Lin

  7. [15]

    Xu He, Zhiyong Fu, Yao Wang, Chang Liu, and Yang Guo. 2022. Accurate timing prediction at placement stage with look-ahead RC network. InProceedings of the 59th ACM/IEEE Design Automation Conference

  8. [16]

    Tsung-Wei Huang and Martin DF Wong. 2015. OpenTimer: A high-performance timing analysis tool. In2015 IEEE/ACM International Conference on Computer- Aided Design (ICCAD)

  9. [17]

    InProceedings of the 59th ACM/IEEE Design Automation Conference (DAC ’22)

    A timing engine inspired graph neural network model for pre-routing slack prediction. InProceedings of the 59th ACM/IEEE Design Automation Conference (DAC ’22)

  10. [18]

    Kahng, Seokhyeong Kang, Hyein Lee, Siddhartha Nath, and Jyoti Wadhwani

    Andrew B. Kahng, Seokhyeong Kang, Hyein Lee, Siddhartha Nath, and Jyoti Wadhwani. 2013. Learning-based approximation of interconnect delay and slew in signoff timing tools. In2013 ACM/IEEE International Workshop on System Level Interconnect Prediction (SLIP)

  11. [19]

    Kahng, Mulong Luo, and Siddhartha Nath

    Andrew B. Kahng, Mulong Luo, and Siddhartha Nath. 2015. SI for free: machine learning of interconnect coupling delay and transition effects. In2015 ACM/IEEE International Workshop on System Level Interconnect Prediction (SLIP)

  12. [20]

    Leilei Jin, Rongliang Fu, Zhen Zhuang, Liang Xiao, Fangzhou Liu, Bei Yu, and Tsung-Yi Ho. 2025. ChronoTE: Crosstalk-Aware Timing Estimation for Rout- ing Optimization via Edge-Enhanced GNNs. In2025 IEEE/ACM International Conference on Computer-Aided Design (ICCAD)

  13. [21]

    Yujia Li, Daniel Tarlow, Marc Brockschmidt, and Richard Zemel. 2015. Gated graph sequence neural networks.arXiv preprint arXiv:1511.05493(2015)

  14. [22]

    Rongjian Liang, Zhiyao Xie, Jinwook Jung, Vishnavi Chauha, Yiran Chen, Jiang Hu, Hua Xiang, and Gi-Joon Nam. 2020. Routing-Free Crosstalk Prediction. In 2020 IEEE/ACM International Conference On Computer Aided Design (ICCAD)

  15. [23]

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

  16. [24]

    Bo Peng, Eric Alcaide, Quentin Anthony, Alon Albalak, Samuel Arcadinho, Stella Biderman, Huanqi Cao, Xin Cheng, Michael Chung, Matteo Grella, et al. 2023. Rwkv: Reinventing rnns for the transformer era.arXiv preprint arXiv:2305.13048 (2023)

  17. [25]

    Pei Quan, Yong Shi, Minglong Lei, Jiaxu Leng, Tianlin Zhang, and Lingfeng Niu

  18. [26]

    Daniela Sánchez Lopera and Wolfgang Ecker. 2022. Applying GNNs to Timing Estimation at RTL. InProceedings of the 41st IEEE/ACM International Conference on Computer-Aided Design (ICCAD ’22). Article 3

  19. [27]

    Prianka Sengupta, Aakash Tyagi, Yiran Chen, and Jiang Hu. 2022. How Good Is Your Verilog RTL Code? A Quick Answer from Machine Learning. In2022 IEEE/ACM International Conference On Computer Aided Design (ICCAD)

  20. [28]

    Yutao Sun, Li Dong, Shaohan Huang, Shuming Ma, Yuqing Xia, Jilong Xue, Jiany- ong Wang, and Furu Wei. 2023. Retentive network: A successor to transformer for large language models.arXiv preprint arXiv:2307.08621(2023)

  21. [29]

    Veronika Thost and Jie Chen. 2021. Directed acyclic graph neural networks. arXiv preprint arXiv:2101.07965(2021)

  22. [30]

    T Konstantin Rusch, Michael M Bronstein, and Siddhartha Mishra. 2023. A survey on oversmoothing in graph neural networks.arXiv preprint arXiv:2303.10993 (2023)

  23. [31]

    Petar Veličković, Guillem Cucurull, Arantxa Casanova, Adriana Romero, Pietro Lio, and Yoshua Bengio. 2017. Graph attention networks.arXiv preprint arXiv:1710.10903(2017)

  24. [32]

    Mingjun Wang, Yihan Wen, Bin Sun, Jianan Mu, Juan Li, Xiaoyi Wang, Jing Justin Ye, Bei Yu, and Huawei Li. [n. d.]. Bridging Layout and RTL: Knowledge Dis- tillation based Timing Prediction. InForty-second International Conference on Machine Learning

  25. [33]

    Ziyi Wang, Fangzhou Liu, Tsung-Yi Ho, David Pan, and Bei Yu. 2025. NUA-Timer: Pre-Synthesis Timing Prediction Under Non-Uniform Input Arrival Times. In 2025 IEEE/ACM International Conference on Computer-Aided Design (ICCAD)

  26. [34]

    Ashish Vaswani, Noam Shazeer, Niki Parmar, Jakob Uszkoreit, Llion Jones, Aidan N Gomez, Łukasz Kaiser, and Illia Polosukhin. 2017. Attention is all you need.Advances in neural information processing systems30 (2017)

  27. [35]

    Qitian Wu, Wentao Zhao, Chenxiao Yang, Hengrui Zhang, Fan Nie, Haitian Jiang, Yatao Bian, and Junchi Yan. 2023. Sgformer: Simplifying and empowering transformers for large-graph representations.Advances in Neural Information Processing Systems36 (2023)

  28. [36]

    Zhiyao Xie, Rongjian Liang, Xiaoqing Xu, Jiang Hu, Chen-Chia Chang, Jingyu Pan, and Yiran Chen. 2022. Preplacement Net Length and Timing Estimation by Customized Graph Neural Network.IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems41, 11 (2022)

  29. [37]

    Zhiyao Xie, Rongjian Liang, Xiaoqing Xu, Jiang Hu, Yixiao Duan, and Yiran Chen

  30. [38]

    Ziyi Wang, Siting Liu, Yuan Pu, Song Chen, Tsung-Yi Ho, and Bei Yu. 2023. Restructure-Tolerant Timing Prediction via Multimodal Fusion. In2023 60th ACM/IEEE Design Automation Conference (DAC)

  31. [39]

    Ceyu Xu, Pragya Sharma, Tianshu Wang, and Lisa Wu Wills. 2023. Fast, Robust and Transferable Prediction for Hardware Logic Synthesis. InProceedings of the 56th Annual IEEE/ACM International Symposium on Microarchitecture (MICRO ’23)

  32. [40]

    Songlin Yang, Bailin Wang, Yu Zhang, Yikang Shen, and Yoon Kim. 2024. Paral- lelizing linear transformers with the delta rule over sequence length.Advances in neural information processing systems37 (2024)

  33. [41]

    Tianzhu Ye, Li Dong, Yuqing Xia, Yutao Sun, Yi Zhu, Gao Huang, and Furu Wei

  34. [42]

    Seongjun Yun, Minbyul Jeong, Raehyun Kim, Jaewoo Kang, and Hyunwoo J Kim

  35. [43]

    Ceyu Xu, Chris Kjellqvist, and Lisa Wu Wills. 2022. SNS’s not a synthesizer: a deep-learning-based synthesis predictor. InProceedings of the 49th Annual International Symposium on Computer Architecture (ISCA ’22)

  36. [44]

    Ruizhe Zhong, Junjie Ye, Zhentao Tang, Shixiong Kai, Mingxuan Yuan, Jianye Hao, and Junchi Yan. 2024. PreRoutGNN for timing prediction with order preserving partition: global circuit pre-training, local delay learning and attentional cell modeling. Article 1905

  37. [49]

    Graph transformer networks.Advances in neural information processing systems32 (2019)

  38. [50]

    Xinyun Zhang, Binwu Zhu, Fangzhou Liu, Ziyi Wang, Peng Xu, Hong Xu, and Bei Yu. 2024. Disentangle, Align and Generalize: Learning A Timing Predictor from Different Technology Nodes. InProceedings of the 61st ACM/IEEE Design Automation Conference (DAC ’24). Article 133

  39. [2019]

    In IEEE/WIC/ACM International Conference on Web Intelligence - Companion Volume (Thessaloniki, Greece)(WI ’19 Companion)

    A Brief Review of Receptive Fields in Graph Convolutional Networks. In IEEE/WIC/ACM International Conference on Web Intelligence - Companion Volume (Thessaloniki, Greece)(WI ’19 Companion)

  40. [2021]

    InProceedings of the 26th Asia and South Pacific Design Automation Conference

    Net2: A graph attention network method customized for pre-placement net length estimation. InProceedings of the 26th Asia and South Pacific Design Automation Conference

  41. [2022]

    InProceedings of the 59th ACM/IEEE Design Automation Conference

    A timing engine inspired graph neural network model for pre-routing slack prediction. InProceedings of the 59th ACM/IEEE Design Automation Conference

  42. [2023]

    A Machine Learning Approach to Improving Timing Consistency between Global Route and Detailed Route.ACM Trans. Des. Autom. Electron. Syst.29, 1, Article 18 (Dec. 2023)

  43. [2024]

    Differential transformer.arXiv preprint arXiv:2410.05258(2024)

Pith tools

Reviewed August 1, 2026 · model on record in the stance chip above.