ScaleRTL: Scaling LLMs with Reasoning Data and Test-Time Compute for Accurate RTL Code Generation

• URL: http://arxiv.org/abs/2506.05566v2
• Date: Tue, 15 Jul 2025 21:44:21 GMT
• Title: ScaleRTL: Scaling LLMs with Reasoning Data and Test-Time Compute for Accurate RTL Code Generation
• Abstract summary: We introduce ScaleRTL, the first reasoning LLM for RTL coding that scales up both high-quality reasoning data and test-time compute.<n>Specifically, we curate a diverse set of long chain-of-thought reasoning traces averaging 56K tokens each, resulting in a dataset of 3.5B tokens that captures rich RTL knowledge.<n>Fine-tuning a general-purpose reasoning model on this corpus yields ScaleRTL that is capable of deep RTL reasoning.
• Score: 4.965247405975508
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Recent advances in large language models (LLMs) have enabled near-human performance on software coding benchmarks, but their effectiveness in RTL code generation remains limited due to the scarcity of high-quality training data. While prior efforts have fine-tuned LLMs for RTL tasks, they do not fundamentally overcome the data bottleneck and lack support for test-time scaling due to their non-reasoning nature. In this work, we introduce ScaleRTL, the first reasoning LLM for RTL coding that scales up both high-quality reasoning data and test-time compute. Specifically, we curate a diverse set of long chain-of-thought reasoning traces averaging 56K tokens each, resulting in a dataset of 3.5B tokens that captures rich RTL knowledge. Fine-tuning a general-purpose reasoning model on this corpus yields ScaleRTL that is capable of deep RTL reasoning. Subsequently, we further enhance the performance of ScaleRTL through a novel test-time scaling strategy that extends the reasoning process via iteratively reflecting on and self-correcting previous reasoning steps. Experimental results show that ScaleRTL achieves state-of-the-art performance on VerilogEval and RTLLM, outperforming 18 competitive baselines by up to 18.4% on VerilogEval and 12.7% on RTLLM.

Related papers

• LLM4RTL: Tool-Assisted LLM for RTL Generation [5.338291061810534]
  Large language models (LLMs) have facilitated impressive progress in software engineering, code generation, tooling, and systems.<n>We propose a judge-renew-check-renew-check'' pipeline which updates a current public dataset using a hierarchy of state-of-the-art commercial LLM models.<n>Our experiments identify some common weaknesses of LLMs in rule-based reasoning and logic, and consequently, in RTL code-generation.
  arXiv  Detail & Related papers  (2026-06-13T23:10:40Z)
• CASS-RTL: Correctness-Aware Subspace Steering for RTL Generation with LLMs [0.0]
  This work proposes CASS-RTL, a first-of-its-kind framework for discovering and leveraging LLMs' correctness-aware components.<n>We observe 10%-20% improvement in pass@1/5/10 accuracy on VerilogEval and 5% improvement on CVDP.
  arXiv  Detail & Related papers  (2026-06-04T04:02:51Z)
• StepPRM-RTL: Stepwise Process-Reward Guided LLM Fine-Tuning for Enhanced RTL Synthesis [8.522674895225597]
  We present StepPRM-RTL, a novel framework that combines stepwise trajectory modeling, process-reward modeling, and retrieval-augmented fine-tuning.<n>StepPRM-RTL constructs stepwise reasoning trajectories from canonical solutions, where each step contains a rationale and incremental code modification.<n>Monte Carlo Tree Search (MCTS) explores alternative reasoning paths, enriching the training dataset with high-quality trajectories.
  arXiv  Detail & Related papers  (2026-06-02T21:52:48Z)
• Unified Data Selection for LLM Reasoning [66.0118347918729]
  Training Large Language Models (LLMs) for complex, long-CoT reasoning is often bottlenecked by the need for massive high-quality reasoning data.<n>We propose High-Entropy Sum (HES), a training-free metric that quantifies reasoning quality by summing only the entropy of the top (e.g., 0.5%) highest-entropy tokens in each reasoning sample.
  arXiv  Detail & Related papers  (2026-05-21T12:21:41Z)
• ACE-RTL: When Agentic Context Evolution Meets RTL-Specialized LLMs [12.204779627626273]
  ACE-RTL integrates an RTL-specialized LLM, trained on a large-scale dataset of 1.7 million RTL samples.<n>On the Comprehensive Verilog Design Problems (CVDP) benchmark, ACE-RTL achieves up to a 44.87% pass rate improvement over 14 competitive baselines.
  arXiv  Detail & Related papers  (2026-02-10T19:09:13Z)
• CVeDRL: An Efficient Code Verifier via Difficulty-aware Reinforcement Learning [57.24524263804788]
  Code verifiers play a critical role in post-verification for LLM-based code generation.<n>Existing supervised fine-tuning methods suffer from data scarcity, high failure rates, and poor inference efficiency.<n>We show that naive RL with only functionality rewards fails to generate effective unit tests for difficult branches and samples.
  arXiv  Detail & Related papers  (2026-01-30T10:33:29Z)
• VeriDispatcher: Multi-Model Dispatching through Pre-Inference Difficulty Prediction for RTL Generation Optimization [15.38010259999926]
  Large Language Models (LLMs) show strong performance in RTL generation, but different models excel on different tasks because of architecture and training differences.<n>We propose VeriDispatcher, a multi-LLM RTL generation framework that dispatches each RTL task to suitable LLMs based on pre-inference difficulty prediction.
  arXiv  Detail & Related papers  (2025-11-27T20:45:26Z)
• Reasoning with Confidence: Efficient Verification of LLM Reasoning Steps via Uncertainty Heads [104.9566359759396]
  We propose a lightweight alternative for step-level reasoning verification based on data-driven uncertainty scores.<n>Our findings suggest that the internal states of LLMs encode their uncertainty and can serve as reliable signals for reasoning verification.
  arXiv  Detail & Related papers  (2025-11-09T03:38:29Z)
• DecoRTL: A Run-time Decoding Framework for RTL Code Generation with LLMs [0.0]
  We show that large language models (LLMs) exhibit low confidence in regions of structural ambiguity or semantic complexity.<n>We introduce DecoRTL, a novel run-time decoding strategy, that is both syntax-aware and contrastive for RTL code generation.<n>Our approach operates entirely at inference time without requiring any additional model fine-tuning.
  arXiv  Detail & Related papers  (2025-07-03T01:17:44Z)
• PRO-V: An Efficient Program Generation Multi-Agent System for Automatic RTL Verification [6.983135183126461]
  Pro-V is a fully program generation multi-agent system for robust RTL verification.<n>It incorporates an efficient best-of-n iterative sampling strategy to enhance the correctness of generated testbenches.<n>Pro-V attains a verification accuracy of 87.17% on golden RTL implementations and 76.28% on RTL mutants.
  arXiv  Detail & Related papers  (2025-06-13T20:06:34Z)
• QiMeng-CodeV-R1: Reasoning-Enhanced Verilog Generation [51.393569044134445]
  Large language models (LLMs) trained via reinforcement learning with verifiable reward (RLVR) have achieved breakthroughs on tasks with explicit, automatable verification.<n> Extending RLVR to automatically generating hardware description languages (HDLs) like Verilog from natural-language (NL) specifications, however, poses three key challenges.<n>We introduce CodeV-R1, an RLVR framework for training Verilog generation LLMs.
  arXiv  Detail & Related papers  (2025-05-30T03:51:06Z)
• RTL++: Graph-enhanced LLM for RTL Code Generation [0.0]
  Traditional register transfer level (RTL) design methods are manual, time-consuming, and prone to errors.<n>Open-source models offer alternatives; however, they frequently fall short in quality/correctness.<n>This paper proposes RTL++, a first-of-its-kind LLM-assisted method for RTL code generation.
  arXiv  Detail & Related papers  (2025-05-11T00:17:26Z)
• Putting the Value Back in RL: Better Test-Time Scaling by Unifying LLM Reasoners With Verifiers [57.95157497749428]
  We propose RL$V$ that augments any value-free'' RL method by jointly training the LLM as both a reasoner and a generative verifier.<n> RL$V$ boosts MATH accuracy by over 20% with parallel sampling and enables $8-32times$ efficient test-time compute scaling.
  arXiv  Detail & Related papers  (2025-05-07T22:41:26Z)
• RTLRepoCoder: Repository-Level RTL Code Completion through the Combination of Fine-Tuning and Retrieval Augmentation [6.428086269916113]
  We propose RTLRepoCoder, a groundbreaking solution that incorporates specific fine-tuning and Retrieval-Augmented Generation (RAG) for repository-level Verilog code completion.<n>Our solution achieves state-of-the-art performance on public benchmark, significantly surpassing GPT-4 and advanced domain-specific LLMs on Edit Similarity and Exact Match rate.
  arXiv  Detail & Related papers  (2025-04-11T09:04:50Z)
• MetRex: A Benchmark for Verilog Code Metric Reasoning Using LLMs [2.0921175288836746]
  Large Language Models (LLMs) have been applied to various hardware design tasks, including Verilog code generation, EDA tool scripting, and RTL bug fixing.<n>In this paper, we assess the ability of LLMs to reason about post-synthesis metrics of Verilog designs.<n>We introduce MetRex, a large-scale dataset comprising 25,868 Verilog HDL designs and their corresponding post-synthesis metrics, namely area, delay, and static power.
  arXiv  Detail & Related papers  (2024-11-05T19:52:58Z)
• Provenance: A Light-weight Fact-checker for Retrieval Augmented LLM Generation Output [49.893971654861424]
  We present a light-weight approach for detecting nonfactual outputs from retrieval-augmented generation (RAG) We compute a factuality score that can be thresholded to yield a binary decision. Our experiments show high area under the ROC curve (AUC) across a wide range of relevant open source datasets.
  arXiv  Detail & Related papers  (2024-11-01T20:44:59Z)
• OriGen:Enhancing RTL Code Generation with Code-to-Code Augmentation and Self-Reflection [54.775409528658486]
  OriGen is a fully open-source framework that incorporates self-reflection capabilities and a novel dataset augmentation methodology. Our approach employs a code-tocode augmentation technique to enhance the quality of open-source RTL code datasets.
  arXiv  Detail & Related papers  (2024-07-23T07:22:25Z)
• ITERTL: An Iterative Framework for Fine-tuning LLMs for RTL Code Generation [9.409062607311528]
  Large language models (LLMs) have demonstrated excellent performance, inspiring researchers to explore their use in automating register transfer level (RTL) code generation.<n>Existing approaches to fine-tune LLMs for RTL generation typically are conducted on fixed datasets.<n>We introduce an iterative training paradigm named ITERTL to mitigate these issues.<n>Our model outperforms GPT4 and state-of-the-art (SOTA) open-source models, achieving remarkable 53.8% pass@1 rate on VerilogEval-human benchmark.
  arXiv  Detail & Related papers  (2024-06-28T01:44:57Z)
• DARG: Dynamic Evaluation of Large Language Models via Adaptive Reasoning Graph [70.79413606968814]
  We introduce Dynamic Evaluation of LLMs via Adaptive Reasoning Graph Evolvement (DARG) to dynamically extend current benchmarks with controlled complexity and diversity. Specifically, we first extract the reasoning graphs of data points in current benchmarks and then perturb the reasoning graphs to generate novel testing data. Such newly generated test samples can have different levels of complexity while maintaining linguistic diversity similar to the original benchmarks.
  arXiv  Detail & Related papers  (2024-06-25T04:27:53Z)
• Temporal Scaling Law for Large Language Models [57.83580734589091]
  We propose the novel concept of Temporal Scaling Law, studying how the test loss of an LLM evolves as the training steps scale up.<n>In contrast to modeling the test loss as a whole in a coarse-grained manner, we break it down and dive into the fine-grained test loss of each token position.<n>We derive the much more precise temporal scaling law by studying the temporal patterns of the parameters in the dynamic hyperbolic-law.
  arXiv  Detail & Related papers  (2024-04-27T05:49:11Z)
• FFN-SkipLLM: A Hidden Gem for Autoregressive Decoding with Adaptive Feed Forward Skipping [49.66872823080736]
  Autoregressive Large Language Models (e.g., LLaMa, GPTs) are omnipresent achieving remarkable success in language understanding and generation. To mitigate overload incurred during generation, several early-exit and layer-dropping strategies have been proposed. We propose FFN-SkipLLM, which is an input-adaptive feed-forward skipping strategy.
  arXiv  Detail & Related papers  (2024-04-05T02:35:43Z)
• Reflection-Tuning: Data Recycling Improves LLM Instruction-Tuning [79.32236399694077]
  Low-quality data in the training set are usually detrimental to instruction tuning. We propose a novel method, termed "reflection-tuning" This approach utilizes an oracle LLM to recycle the original training data by introspecting and enhancing the quality of instructions and responses in the data.
  arXiv  Detail & Related papers  (2023-10-18T05:13:47Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.