Pith. sign in

REVIEW 4 major objections 6 minor 36 references

Free and Fair Hardware: A Pathway to Copyright Infringement-Free Verilog Generation using LLMs

T0 review · 4 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 This paper argues that filtering Verilog training data for open licenses and copyright markers nearly eliminates a fine-tuned model's tendency to reproduce protected code, while still improving code-generation accuracy over its base model.

desk verdict A genuinely useful copyright-filtered Verilog dataset and a first-of-its-kind infringement benchmark, but the 'infringement-free' claim outruns the evidence. read the letter →

arxiv 2505.06096 v2 pith:OW2AG3RK submitted 2025-05-09 cs.AI

Sam Bush , Matthew DeLorenzo , Phat Tieu , Jeyavijayan Rajendran This is my paper · ORCID
classification cs.AI
keywords LLMcopyrightVeriloggenerationhardwaredesigndatasetcurationlicensefilteringfine-tuningintellectualpropertyRTLcode
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 argues that copyright risk in Verilog-generating LLMs comes largely from uncurated training data, and that a dataset filtered for open licenses and copyright markers can nearly eliminate that risk without sacrificing code quality. To show this, the authors assemble FreeSet, 222,624 Verilog files from GitHub repositories that carry open-source licenses, with files containing words like 'proprietary', 'confidential', or 'all rights reserved' removed. They then fine-tune Llama-3.1-8B-Instruct on FreeSet to produce FreeV, reporting a 3% copyright violation rate on their similarity benchmark, the lowest among compared models, alongside a 10.1 percentage-point improvement in VerilogEval Human pass@10 over the base model. The paper also contributes the benchmark itself: prompting models with the first 20% of copyrighted Verilog files and flagging outputs whose cosine similarity exceeds 0.8.

What carries the argument

The load-bearing machinery is the dataset curation pipeline. It combines GitHub API queries granularized by creation date and license to collect about 1.3 million Verilog files, repository-level license allowlisting (MIT, Apache 2.0, GPL, LGPL, Mozilla, Creative Commons, Eclipse, BSD), file-by-file header scanning for 'proprietary', 'confidential', and 'all rights reserved' (which removed over 2,000 protected files), MinHash-based Locality Sensitive Hashing duplication removal with a 0.85 similarity threshold, and syntax verification with Icarus Verilog. The evaluation pipeline is equally central: a copyright benchmark that prompts models with the first 20% (64 words) of each of 100 protected modules and counts a violation when a generated output has cosine similarity at least 0.8 to the protected code.

What would settle it

Scan every file in FreeSet for copyright notices that do not contain any of the filtered keywords (for example, a bare '© 2020 Example Corp' header); if even one such file exists, the dataset's copyright-cleanness claim is not established. Alternatively, run the paper's benchmark with full-file prompts on FreeV and check whether the violation rate exceeds 3%.

Watch

Extended reading notes

Core claim

The central claim is that a hardware-code LLM can be fine-tuned on a large dataset scrubbed of copyrighted material, and that this scrubbing is the reason the fine-tuned model FreeV reproduces protected designs only 3% of the time, versus 9-15% for prior Verilog-tuned models. The scrubbing operates in two layers: repository-level license screening that keeps only repositories under recognized open-source licenses, and file-level scanning of header comments for copyright markers. The paper further claims that this safer training does not cost functionality: FreeV improves over its base model on VerilogEval Human, with pass@10 rising from 25.9% to 36.0%, though it remains below state-of-the-art Verilog-tuned models that also use instruction tuning.

Load-bearing premise

The filter catches copyrighted files only when they carry the scanned keywords or a recognizable open-source license marker, so protected code that lacks 'proprietary', 'confidential', or 'all rights reserved' in its header could remain in FreeSet and make the 3% violation rate an underestimate.

Editorial extensions

If this is right

  • FreeSet's license and copyright filters can serve as a template for future hardware datasets; it is the largest open-source Verilog dataset reported, at 16.5 GB and 222,624 files.
  • Fine-tuning on copyright-cleaned data does not inherently reduce functional performance: FreeV's pass@5 and pass@10 improve by 7.9 and 10.1 percentage points over its base.
  • The benchmark can be applied to other Verilog-tuned models to estimate their copyright risk before deployment.
  • Because 3% of FreeV's outputs still clear the similarity threshold, even a carefully filtered dataset and model retain some infringement risk.

Reading between the lines

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

  • The curation pipeline likely transfers to other hardware description languages such as SystemVerilog and VHDL, and to software code domains, whenever header comments are the primary copyright markers.
  • The benchmark's design (first 20% of a file, 64-word prompt) probably undercounts memorization; prompting with larger fractions or with the model's own completions of long files could raise violation rates for all models, including FreeV.
  • Repository-level license filtering excludes unlicensed repositories entirely, which may discard legally reusable code (e.g., code placed in the public domain without a formal license); a provenance-aware filter could recover some of that data.
  • The paper's own suggestion to use structural similarity measures like GNN4IP hints that text-cosine benchmarks miss designs that are structurally copied but textually rewritten.
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

4 major / 6 minor

Summary. The paper proposes an end-to-end pipeline for building a copyright-filtered Verilog training dataset (FreeSet, over 220k files), a cosine-similarity benchmark for estimating how often Verilog-tuned LLMs reproduce protected code, and a continually pre-trained model (FreeV, based on Llama-3.1-8B-Instruct). The main reported results are that FreeV has a 3% copyright-violation rate, lower than prior fine-tuned Verilog models, and that it improves VerilogEval-Human pass@10 from 25.9% to 36.0% compared with its 4-bit base model.

Significance. If the claims were fully established, the contributions would be useful: FreeSet appears to be the largest open Verilog corpus at 16.5 GB / 222,624 files, the curation pipeline addresses a real gap in prior datasets, and the VerilogEval gains are concrete and measured rather than asserted. The authors also release the dataset and describe their filtering framework, which supports reproducibility. The main significance is therefore in the dataset and pipeline, not in the legal claim: the 3% violation rate is a preliminary estimate on a narrow benchmark, and the paper itself acknowledges in Section V that the benchmark is preliminary. The central safety claim, however, is currently stated much more strongly than the evidence supports.

major comments (4)
  1. [Section IV-B and Section III-A] The headline 3% violation rate is not statistically grounded as presented. The benchmark uses only 100 prompts drawn from a 2K-file corpus, and a 3/100 result has a 95% binomial confidence interval of roughly 0.6% to 8.5%, which overlaps plausible rates for several other models and for the base model. The paper does not specify how cosine similarity is computed (token-level? embedding? over which fields?), how the 20% file prefixes are truncated to 64 words, or whether the 100 prompts are independent samples. Please report the full experimental protocol, include confidence intervals or exact per-model counts, and soften the claim 'smallest potential for hardware copyright infringement' accordingly.
  2. [Section III-C.2 and Section IV-B] The filtering pipeline is under-inclusive for the strong claim that FreeSet is free of copyright-protected material. The file-by-file check inspects only header comments for a fixed keyword list ('proprietary', 'confidential', 'all rights reserved'), and the benchmark's test corpus is built from files that carry detectable copyright declarations. This means the 3% rate estimates leakage of the same class of explicitly marked files that the filter is designed to remove; it does not estimate leakage of protected Verilog whose notices appear in the body, contain no such keywords, or have no notice at all. The paper itself reports finding over 2K proprietary files inside reportedly open-source repositories, which shows that repository-level licenses are insufficient, but it does not validate the header-keyword filter against unmarked protected code. Please either add a validation study (e.g., manual or provenance-based audit of a random sample of surviving files) or explicitly reframe the claim as a reduction in risk for detected, explicitly marked copyrighted files.
  3. [Section V and Abstract] The paper's own Discussion section calls the benchmark 'preliminary' and lists open problems such as additional prompting methods and expanded test corpora, yet the Abstract and Conclusion state as a result that FreeV has 'the smallest potential for hardware copyright infringement among prior works.' Since the benchmark cannot currently establish a legal or even a robust empirical infringement rate, the abstract and conclusion should be aligned with the provisional nature of the benchmark. At minimum, 'infringement-free' and 'smallest risk' should be replaced with a statement about measured similarity-based violation rates on the proposed benchmark.
  4. [Section III-A and Section IV-A] The construction of the 2K-file copyrighted benchmark corpus is not described precisely. The text says this dataset is 'collected in Section III-B,' but Section III-B describes the general FreeSet curation pipeline, not the curation of the copyrighted benchmark set. It is also unclear whether the 100 benchmark prompts are drawn from the same 2K proprietary files found during FreeSet filtering or from a separate corpus, and how the modules are isolated after comment stripping. Please clarify the provenance of the benchmark files, whether they overlap with files removed from FreeSet, and how the 100 prompts were selected from the 2K files; without this, the benchmark's coverage and independence cannot be assessed.
minor comments (6)
  1. [Section III-C.2] There is a typo in 'only epositories with the fair-use modules are utilized'; it should read 'only repositories.'
  2. [Section VI] The concluding paragraph states that the curation framework 'results in a curated and open-source dataset FreeV'; the dataset is named FreeSet, not FreeV, throughout the rest of the paper.
  3. [References] Reference [12] has a typo in the title: 'Cerilog' should be 'Verilog.'
  4. [Section III-A] The phrase 'with a limit of 64 words per prompt' is ambiguous when combined with 'the first 20% of a copyrighted code file'; please specify whether the 20% prefix is truncated to 64 words or whether 64 words is a separate cap, and report the distribution of prompt lengths.
  5. [Section I and III-B] The paper uses 'fair-use data' and 'fair-use modules' to describe the dataset, but fair use is a legal defense, not a property of a corpus. This wording should be changed to 'properly licensed' or 'reusable' to avoid implying a legal determination that the paper does not make.
  6. [Table I and dataset link] The dataset is described as open-source, but the Hugging Face link contains the suffix 'LabUse'; if the dataset is restricted to lab or research use, the 'open-source' characterization should be qualified in Table I and the abstract.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the violation-rate benchmark is a held-out corpus and the VerilogEval gains are measured against an external benchmark.

full rationale

The paper's central claims are (a) FreeV has a 3% copyright-violation rate on a benchmark of copyrighted Verilog files, and (b) FreeV improves VerilogEval Human pass@10 from 25.9% to 36.0% over its base model. Neither claim reduces to its own inputs by construction. The violation rate is measured on a separately curated 2K-file corpus of files with copyright declarations (Section III-A), while FreeSet was built by removing files with open-license and header-keyword filters (Section III-C.2); the test files are a holdout set, not a fitted target. The cosine-similarity threshold of 0.8 is stated a priori and is not tuned to produce the authors' own low rate. Moreover, the result is not forced: the base Llama-3.1 model already shows a low violation rate, and fine-tuning on FreeSet increases it by 1% to 3%, so the measured 3% is an empirical outcome rather than an identity. The pass@k improvements are evaluated on the external VerilogEval-Human benchmark and are not used in training. The paper's dependencies on prior work (VeriGen for de-duplication and evaluation protocol) are external and non-load-bearing, and there is no self-citation chain, imported uniqueness theorem, or ansatz smuggled in via citation. The skeptical concern that unmarked copyrighted code could survive the header-keyword filter is a legitimate threat to the strength of the 'infringement-free' framing, but that is a correctness/external-validity limitation, not a circularity: the paper itself calls the benchmark 'preliminary' and lists expanding prompting methods and similarity metrics as future work in Section V. No circular step meeting the quoted-evidence bar was found.

Assumptions & free parameters 6 free parameters · 3 assumptions · 0 invented entities

The central claims rely on a handful of hand-chosen thresholds (similarity cutoff, prompt design, keyword list) and on unvalidated legal and measurement assumptions. No new physical or formal entities are introduced. The dataset and model are engineering artifacts.

free parameters (6)
  • Copyright similarity threshold = 0.8
    Chosen in Section III-A as the definition of a violation; directly determines all reported violation rates, including the 3% headline.
  • Prompt truncation for copyright benchmark = first 20% of file, max 64 words
    Section III-A; hand-picked prompt construction that affects whether models retrieve memorized code.
  • Benchmark prompt count = 100
    Section III-A; no power analysis; the 3% rate is 3 out of 100 prompts, and the 3% vs 2% difference is one prompt.
  • Copyright filter keyword list = proprietary, confidential, all rights reserved, etc.
    Section III-C; the dataset safety guarantee depends on this hand-selected list catching actual copyright notices.
  • De-duplication similarity threshold = 0.85
    Section III-D2; inherited from VeriGen; changes dataset composition and could remove or keep unique copyrighted files.
  • LoRA rank and alpha = 8 and 8
    Section III-E1; training hyperparameters that affect model capacity and downstream VerilogEval results, though not the copyright benchmark directly.
assumptions (3)
  • domain assumption Verilog files from repositories with permissive or non-permissive open-source licenses are safe for training reuse.
    Section III-C; legal assumption that license presence resolves reuse rights; no legal analysis is given.
  • ad hoc to paper Verilog files lacking header-comment keywords such as 'proprietary' or 'confidential' are not copyright-protected.
    Section III-C; this is the core filtering rule and is untested against files that carry copyright without these keywords.
  • ad hoc to paper Cosine similarity of generated output to a copyrighted corpus is a valid proxy for copyright infringement.
    Section III-A; similarity is neither sufficient nor necessary for legal infringement; the paper assumes the 0.8 threshold separates the two.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Free and Fair Hardware: A Pathway to Copyright Infringement-Free Verilog Generation using LLMs." pith.science (2026). https://pith.science/paper/OW2AG3RK

@misc{pith2026250506096,
  author       = {Pith},
  title        = {Pith review of: Free and Fair Hardware: A Pathway to Copyright Infringement-Free Verilog Generation using LLMs},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/OW2AG3RK}},
  note         = {Machine review of arXiv:2505.06096}
}
read the original abstract

Limitations in Large Language Model (LLM) capabilities for hardware design tasks, such as generating functional Verilog codes, have motivated various fine-tuning optimizations utilizing curated hardware datasets from open-source repositories. However, these datasets remain limited in size and contain minimal checks on licensing for reuse, resulting in potential copyright violations by fine-tuned LLMs. Therefore, we propose an evaluation benchmark to estimate the risk of Verilog-trained LLMs to generate copyright-protected codes. To minimize this risk, we present an open-source Verilog dataset, FreeSet, containing over 220k files, along with the automated dataset curation framework utilized to provide additional guarantees of fair-use Verilog data. We then execute an LLM fine-tuning framework consisting of continual pre-training, resulting in a fine-tuned Llama model for Verilog, FreeV. Our results indicate that FreeV demonstrates the smallest risk of copyright-infringement among prior works, with only a 3% violation rate. Furthermore, experimental results demonstrate improvements in Verilog generation functionality over its baseline model, improving VerilogEval pass@10 rates by over 10%.

Figures

Figures reproduced from arXiv: 2505.06096 by the authors.

Figure 1
Figure 1. Framework - Dataset Curation and Continual Pre-Training [PITH_FULL_IMAGE:figures/full_fig_p002_1.png] view at source ↗
Figure 2
Figure 2. Distribution in Verilog File Lengths of FreeSet vs. VeriGen [12] [PITH_FULL_IMAGE:figures/full_fig_p003_2.png] view at source ↗
Figure 3
Figure 3. Hardware Copyright Infringement Rates across LLMs [PITH_FULL_IMAGE:figures/full_fig_p005_3.png] view at source ↗

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

36 extracted references · 17 canonical work pages

  1. [1]

    Attention Is All You Need,

    A. Vaswani, N. Shazeer, N. Parmar, J. Uszkoreit, L. Jones, A. N. Gomez, L. Kaiser, and I. Polosukhin, “Attention Is All You Need,” 2023

    work page 2023

  2. [2]

    Transformers: State- of-the-Art Natural Language Processing,

    T. Wolf, L. Debut, V . Sanh, J. Chaumond, C. Delangue, A. Moi, P. Cistac, T. Rault, R. Louf, M. Funtowicz et al., “Transformers: State- of-the-Art Natural Language Processing,” in Proceedings of the 2020 conference on empirical methods in natural language processing: system demonstrations, 2020, pp. 38–45

    work page 2020

  3. [3]

    12 Practical Large Language Model (LLM) Applications,

    Tim Keary, “12 Practical Large Language Model (LLM) Applications,” https://www.techopedia.com/ 12-practical-large-language-model-llm-applications, 2023, [Online; last accessed 21-Nov-2023]

    work page 2023

  4. [4]

    DeepSeek-Coder: When the Large Language Model Meets Programming–The Rise of Code Intelligence,

    D. Guo, Q. Zhu, D. Yang, Z. Xie, K. Dong, W. Zhang, G. Chen, X. Bi, Y . Wu, Y . Liet al., “DeepSeek-Coder: When the Large Language Model Meets Programming–The Rise of Code Intelligence,” arXiv preprint arXiv:2401.14196, 2024

    arXiv 2024

  5. [5]

    Evaluating Large Language Models Trained on Code,

    M. Chen, J. Tworek, H. Jun, Q. Yuan, H. P. D. O. Pinto, J. Kaplan, H. Edwards, Y . Burda, N. Joseph, G. Brockmanet al., “Evaluating Large Language Models Trained on Code,” arXiv preprint arXiv:2107.03374 , 2021

    arXiv 2021

  6. [6]

    DeepSeek-Coder-V2: Breaking the Barrier of Closed- Source Models in Code Intelligence,

    Q. Zhu, D. Guo, Z. Shao, D. Yang, P. Wang, R. Xu, Y . Wu, Y . Li, H. Gao, S. Ma et al. , “DeepSeek-Coder-V2: Breaking the Barrier of Closed- Source Models in Code Intelligence,” arXiv preprint arXiv:2406.11931, 2024

    arXiv 2024

  7. [7]

    The Stack: 3 TB of permissively licensed source code,

    D. Kocetkov, R. Li, L. B. Allal, J. Li, C. Mou, C. M. Ferrandis, Y . Jernite, M. Mitchell, S. Hughes, T. Wolf et al., “The Stack: 3 TB of permissively licensed source code,” arXiv preprint arXiv:2211.15533 , 2022

    arXiv 2022

  8. [8]

    An Introduction to Microsoft Copilot,

    J. Stratton, “An Introduction to Microsoft Copilot,” in Copilot for Microsoft 365: Harness the Power of Generative AI in the Microsoft Apps You Use Every Day . Springer, 2024, pp. 19–35

    work page 2024

Show all 36 references
  1. [9]

    Code Llama: Open Foundation Models for Code,

    B. Roziere, J. Gehring, F. Gloeckle, S. Sootla, I. Gat, X. E. Tan, Y . Adi, J. Liu, T. Remez, J. Rapin et al., “Code Llama: Open Foundation Models for Code,” arXiv preprint arXiv:2308.12950 , 2023

    2023 arXiv

  2. [10]

    GPT-4 Technical Report,

    J. Achiam, S. Adler, S. Agarwal, L. Ahmad, I. Akkaya, F. L. Aleman, D. Almeida, J. Altenschmidt, S. Altman, S. Anadkat et al. , “GPT-4 Technical Report,” arXiv preprint arXiv:2303.08774 , 2023

    2023 arXiv

  3. [11]

    LLM4EDA: Emerging Progress in Large Language Models for Electronic Design Automation,

    R. Zhong, X. Du, S. Kai, Z. Tang, S. Xu, H.-L. Zhen, J. Hao, Q. Xu, M. Yuan, and J. Yan, “LLM4EDA: Emerging Progress in Large Language Models for Electronic Design Automation,” arXiv preprint arXiv:2401.12224, 2023

    2023 arXiv

  4. [12]

    VeriGen: A Large Language Model for Cerilog Code Generation,

    S. Thakur, B. Ahmad, H. Pearce, B. Tan, B. Dolan-Gavitt, R. Karri, and S. Garg, “VeriGen: A Large Language Model for Cerilog Code Generation,” ACM Transactions on Design Automation of Electronic Systems, vol. 29, no. 3, pp. 1–31, 2024

    2024

  5. [13]

    Chip-Chat: Chal- lenges and Opportunities in Conversational Hardware Design,

    J. Blocklove, S. Garg, R. Karri, and H. Pearce, “Chip-Chat: Chal- lenges and Opportunities in Conversational Hardware Design,” in 2023 ACM/IEEE 5th Workshop on Machine Learning for CAD (MLCAD) . IEEE, 2023, pp. 1–6

    2023

  6. [14]

    ChipGPT: How far are we from natural language hardware design,

    K. Chang, Y . Wang, H. Ren, M. Wang, S. Liang, Y . Han, H. Li, and X. Li, “ChipGPT: How far are we from natural language hardware design,” arXiv preprint arXiv:2305.14019 , 2023

    2023

  7. [15]

    CodeV: Empowering LLMs for Verilog Generation through Multi-Level Summarization,

    Y . Zhao, D. Huang, C. Li, P. Jin, Z. Nan, T. Ma, L. Qi, Y . Pan, Z. Zhang, R. Zhang et al. , “CodeV: Empowering LLMs for Verilog Generation through Multi-Level Summarization,” arXiv preprint arXiv:2407.10424, 2024

    2024 arXiv

  8. [16]

    RTLCoder: Outperforming GPT-3.5 in Design RTL Generation with Our Open- Source Dataset and Lightweight Solution,

    S. Liu, W. Fang, Y . Lu, Q. Zhang, H. Zhang, and Z. Xie, “RTLCoder: Outperforming GPT-3.5 in Design RTL Generation with Our Open- Source Dataset and Lightweight Solution,” in 2024 IEEE LLM Aided Design Workshop (LAD). IEEE, 2024, pp. 1–5

    2024

  9. [17]

    BetterV: Con- trolled Verilog Generation with Discriminative Guidance,

    Z. Pei, H.-L. Zhen, M. Yuan, Y . Huang, and B. Yu, “BetterV: Con- trolled Verilog Generation with Discriminative Guidance,”arXiv preprint arXiv:2402.03375, 2024

    2024 arXiv

  10. [18]

    Large Language Model for Verilog Generation with Golden Code Feedback,

    N. Wang, B. Yao, J. Zhou, X. Wang, Z. Jiang, and N. Guan, “Large Language Model for Verilog Generation with Golden Code Feedback,” arXiv preprint arXiv:2407.18271 , 2024

    2024 arXiv

  11. [19]

    AutoChip: Automating HDL Generation Using LLM Feedback,

    S. Thakur, J. Blocklove, H. Pearce, B. Tan, S. Garg, and R. Karri, “AutoChip: Automating HDL Generation Using LLM Feedback,” arXiv preprint arXiv:2311.04887, 2023

    2023 arXiv

  12. [20]

    RTLFixer: Automatically Fixing RTL Syntax Errors with Large Language Model,

    Y . Tsai, M. Liu, and H. Ren, “RTLFixer: Automatically Fixing RTL Syntax Errors with Large Language Model,” in Proceedings of the 61st ACM/IEEE Design Automation Conference , 2024, pp. 1–6

    2024

  13. [21]

    OpenAI, Microsoft want court to toss lawsuit accusing them of abusing open-source code,

    B. Brittain, “OpenAI, Microsoft want court to toss lawsuit accusing them of abusing open-source code,” Reuters, Jan

  14. [22]

    RTL Hardware IP Protection Using Key-Based Control and Data Flow Obfuscation,

    R. S. Chakraborty and S. Bhunia, “RTL Hardware IP Protection Using Key-Based Control and Data Flow Obfuscation,” in 2010 23rd Interna- tional Conference on VLSI Design , 2010, pp. 405–410

    2010

  15. [23]

    Copyright Traps for Large Language Models,

    M. Meeus, I. Shilov, M. Faysse, and Y .-A. de Montjoye, “Copyright Traps for Large Language Models,” arXiv preprint arXiv:2402.09363 , 2024

    2024 arXiv

  16. [24]

    Beyond Fair Use: Legal Risk Evaluation for Training LLMs on Copyrighted Text,

    N. Rahman and E. Santacana, “Beyond Fair Use: Legal Risk Evaluation for Training LLMs on Copyrighted Text,” in ICML Workshop on Generative AI and Law , 2023

    2023

  17. [25]

    Ongoing lawsuits could affect everyone who uses generative AI,

    P. Samuelson, “Ongoing lawsuits could affect everyone who uses generative AI,” Science, vol. 381, p. 6654, 2023

    2023

  18. [26]

    Ex- ploring Memorization and Copyright Violation in Frontier LLMs: A Study of the New York Times v. OpenAI 2023 Lawsuit,

    J. Freeman, C. Rippe, E. Debenedetti, and M. Andriushchenko, “Ex- ploring Memorization and Copyright Violation in Frontier LLMs: A Study of the New York Times v. OpenAI 2023 Lawsuit,” in Neurips Safe Generative AI Workshop 2024 , 2024

    2023

  19. [27]

    Digger: Detecting copyright content mis-usage in large language model training,

    H. Li, G. Deng, Y . Liu, K. Wang, Y . Li, T. Zhang, Y . Liu, G. Xu, G. Xu, and H. Wang, “Digger: Detecting copyright content mis-usage in large language model training,” arXiv preprint arXiv:2401.00676 , 2024

    2024 arXiv

  20. [28]

    SHIELD: Evaluation and Defense Strategies for Copyright Compliance in LLM Text Generation,

    X. Liu, T. Sun, T. Xu, F. Wu, C. Wang, X. Wang, and J. Gao, “SHIELD: Evaluation and Defense Strategies for Copyright Compliance in LLM Text Generation,” arXiv preprint arXiv:2406.12975 , 2024

    2024 arXiv

  21. [29]

    RTLLM: An Open-Source Benchmark for Design RTL Generation with Large Language Model,

    Y . Lu, S. Liu, Q. Zhang, and Z. Xie, “RTLLM: An Open-Source Benchmark for Design RTL Generation with Large Language Model,” in 2024 29th Asia and South Pacific Design Automation Conference (ASP-DAC). IEEE, 2024, pp. 722–727

    2024

  22. [30]

    LLM-assisted Generation of Hardware Assertions,

    R. Kande, H. Pearce, B. Tan, B. Dolan-Gavitt, S. Thakur, R. Karri, and J. Rajendran, “LLM-assisted Generation of Hardware Assertions,” arXiv preprint arXiv:2306.14027, 2023

    2023 arXiv

  23. [31]

    OpenLLM-RTL: Open Dataset and Benchmark for LLM-Aided Design RTL Generation,

    S. Liu, Y . Lu, W. Fang, M. Li, and Z. Xie, “OpenLLM-RTL: Open Dataset and Benchmark for LLM-Aided Design RTL Generation,” in Proceedings of the 43rd IEEE/ACM International Conference on Computer-Aided Design, 2024, pp. 1–9

    2024

  24. [32]

    Make Every Move Count: LLM-based High-Quality RTL Code Generation Using MCTS,

    M. DeLorenzo, A. B. Chowdhury, V . Gohil, S. Thakur, R. Karri, S. Garg, and J. Rajendran, “Make Every Move Count: LLM-based High-Quality RTL Code Generation Using MCTS,” arXiv preprint arXiv:2402.03289, 2024

    2024 arXiv

  25. [33]

    CraftRTL: High-quality Synthetic Data Generation for Verilog Code Models with Correct-by- Construction Non-Textual Representations and Targeted Code Repair,

    M. Liu, Y .-D. Tsai, W. Zhou, and H. Ren, “CraftRTL: High-quality Synthetic Data Generation for Verilog Code Models with Correct-by- Construction Non-Textual Representations and Targeted Code Repair,” arXiv preprint arXiv:2409.12993 , 2024

    2024 arXiv

  26. [34]

    OriGen: Enhancing RTL Code Generation with Code-to-Code Augmentation and Self-Reflection,

    F. Cui, C. Yin, K. Zhou, Y . Xiao, G. Sun, Q. Xu, Q. Guo, D. Song, D. Lin, X. Zhang et al. , “OriGen: Enhancing RTL Code Generation with Code-to-Code Augmentation and Self-Reflection,” arXiv preprint arXiv:2407.16237, 2024

    2024 arXiv

  27. [35]

    GNN4IP: Graph Neural Network for Hardware Intellectual Property Piracy Detec- tion,

    R. Yasaei, S.-Y . Yu, E. K. Naeini, and M. A. Al Faruque, “GNN4IP: Graph Neural Network for Hardware Intellectual Property Piracy Detec- tion,” in 2021 58th ACM/IEEE Design Automation Conference (DAC) . IEEE, 2021, pp. 217–222

    2021

  28. [2023]

    Available: https://www.reuters.com/legal/litigation/ openai-microsoft-want-court-toss-lawsuit-accusing-them-abusing-open-source-code-2023-01-27/

    [Online]. Available: https://www.reuters.com/legal/litigation/ openai-microsoft-want-court-toss-lawsuit-accusing-them-abusing-open-source-code-2023-01-27/

    2023

Pith tools

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