@inproceedings{huang-etal-2026-circa,
title = "{C}ir{CA}: Aligning Credit Assignment with Circuit Semantics via Atomic Logic Blocks for {RTL} Generation",
author = "Huang, Lei and
Zhang, Rui and
Guo, Jiaming and
Zhang, Yang and
Huang, Di and
Du, Zidong and
Hu, Xing and
Guo, Qi and
Chen, Yunji",
editor = "Lee, Hung-Yi and
Dev, Sunipa and
Li, Jessy and
Elliott, Desmond",
booktitle = "Proceedings of the 2026 Conference on Empirical Methods in Natural Language Processing",
month = oct,
year = "2026",
address = "Budapest, Hungary",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2026.emnlp-main.1450/",
pages = "33112--33128",
ISBN = "979-8-89176-459-0",
abstract = "Reinforcement learning (RL) post-training has emerged as a promising paradigm for LLM-based Verilog generation. However, existing RL credit assignment methods are either overly coarse (sequence-level GRPO) or fragmented (line-level methods), leading to severe misattribution as they ignore the characteristics of Verilog designs that are fundamentally structured through concurrent constructs and procedural control flows. To bridge this gap, we propose CirCA, a novel RL framework that performs credit assignment based on Atomic Logic Blocks (ALBs), the smallest semantically complete units of circuit logic. CirCA decomposes Verilog into ALBs, derives reliable rewards from signal-level feedback, and then constructs logic-enhanced signal dependency cones to localize erroneous ALBs, whose localized responsibilities are further incorporated into GRPO through asymmetric advantage correction. Extensive experiments show CirCA establishes a new state-of-the-art across multiple benchmarks among non-reasoning models, and even surpasses some cutting-edge reasoning models on complex tasks."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="huang-etal-2026-circa">
<titleInfo>
<title>CirCA: Aligning Credit Assignment with Circuit Semantics via Atomic Logic Blocks for RTL Generation</title>
</titleInfo>
<name type="personal">
<namePart type="given">Lei</namePart>
<namePart type="family">Huang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Rui</namePart>
<namePart type="family">Zhang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Jiaming</namePart>
<namePart type="family">Guo</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Yang</namePart>
<namePart type="family">Zhang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Di</namePart>
<namePart type="family">Huang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Zidong</namePart>
<namePart type="family">Du</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Xing</namePart>
<namePart type="family">Hu</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Qi</namePart>
<namePart type="family">Guo</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Yunji</namePart>
<namePart type="family">Chen</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2026-10</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Proceedings of the 2026 Conference on Empirical Methods in Natural Language Processing</title>
</titleInfo>
<name type="personal">
<namePart type="given">Hung-Yi</namePart>
<namePart type="family">Lee</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Sunipa</namePart>
<namePart type="family">Dev</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Jessy</namePart>
<namePart type="family">Li</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Desmond</namePart>
<namePart type="family">Elliott</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Budapest, Hungary</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
<identifier type="isbn">979-8-89176-459-0</identifier>
</relatedItem>
<abstract>Reinforcement learning (RL) post-training has emerged as a promising paradigm for LLM-based Verilog generation. However, existing RL credit assignment methods are either overly coarse (sequence-level GRPO) or fragmented (line-level methods), leading to severe misattribution as they ignore the characteristics of Verilog designs that are fundamentally structured through concurrent constructs and procedural control flows. To bridge this gap, we propose CirCA, a novel RL framework that performs credit assignment based on Atomic Logic Blocks (ALBs), the smallest semantically complete units of circuit logic. CirCA decomposes Verilog into ALBs, derives reliable rewards from signal-level feedback, and then constructs logic-enhanced signal dependency cones to localize erroneous ALBs, whose localized responsibilities are further incorporated into GRPO through asymmetric advantage correction. Extensive experiments show CirCA establishes a new state-of-the-art across multiple benchmarks among non-reasoning models, and even surpasses some cutting-edge reasoning models on complex tasks.</abstract>
<identifier type="citekey">huang-etal-2026-circa</identifier>
<location>
<url>https://aclanthology.org/2026.emnlp-main.1450/</url>
</location>
<part>
<date>2026-10</date>
<extent unit="page">
<start>33112</start>
<end>33128</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T CirCA: Aligning Credit Assignment with Circuit Semantics via Atomic Logic Blocks for RTL Generation
%A Huang, Lei
%A Zhang, Rui
%A Guo, Jiaming
%A Zhang, Yang
%A Huang, Di
%A Du, Zidong
%A Hu, Xing
%A Guo, Qi
%A Chen, Yunji
%Y Lee, Hung-Yi
%Y Dev, Sunipa
%Y Li, Jessy
%Y Elliott, Desmond
%S Proceedings of the 2026 Conference on Empirical Methods in Natural Language Processing
%D 2026
%8 October
%I Association for Computational Linguistics
%C Budapest, Hungary
%@ 979-8-89176-459-0
%F huang-etal-2026-circa
%X Reinforcement learning (RL) post-training has emerged as a promising paradigm for LLM-based Verilog generation. However, existing RL credit assignment methods are either overly coarse (sequence-level GRPO) or fragmented (line-level methods), leading to severe misattribution as they ignore the characteristics of Verilog designs that are fundamentally structured through concurrent constructs and procedural control flows. To bridge this gap, we propose CirCA, a novel RL framework that performs credit assignment based on Atomic Logic Blocks (ALBs), the smallest semantically complete units of circuit logic. CirCA decomposes Verilog into ALBs, derives reliable rewards from signal-level feedback, and then constructs logic-enhanced signal dependency cones to localize erroneous ALBs, whose localized responsibilities are further incorporated into GRPO through asymmetric advantage correction. Extensive experiments show CirCA establishes a new state-of-the-art across multiple benchmarks among non-reasoning models, and even surpasses some cutting-edge reasoning models on complex tasks.
%U https://aclanthology.org/2026.emnlp-main.1450/
%P 33112-33128
Markdown (Informal)
[CirCA: Aligning Credit Assignment with Circuit Semantics via Atomic Logic Blocks for RTL Generation](https://aclanthology.org/2026.emnlp-main.1450/) (Huang et al., EMNLP 2026)
ACL
- Lei Huang, Rui Zhang, Jiaming Guo, Yang Zhang, Di Huang, Zidong Du, Xing Hu, Qi Guo, and Yunji Chen. 2026. CirCA: Aligning Credit Assignment with Circuit Semantics via Atomic Logic Blocks for RTL Generation. In Proceedings of the 2026 Conference on Empirical Methods in Natural Language Processing, pages 33112–33128, Budapest, Hungary. Association for Computational Linguistics.