<p>量子運輸公司的獨特解碼器在高性能雲環境中取得了新的進展</p>
<p>以色列拉阿納納，2026 年 2 月 24 日（全球新聞通訊社）——Rail Vision Ltd.（納斯達克：RVSN）（“Rail Vision” 或 “公司”），一家專注於轉變鐵路安全和鐵路數據市場的早期商業化技術公司，今天宣佈其控股子公司 Quantum Transportation Ltd.（“Quantum Transportation”），一家量子計算創新公司，已成功在 AWS 雲上實施其基於變壓器的神經解碼器，這標誌着在運輸領域實現現實世界量子應用的重要里程碑。</p>
<p>在最近推出的變壓器神經解碼器的基礎上，該解碼器在模擬中超越了經典量子錯誤校正（QEC）算法，並交付了其第一個通用錯誤校正原型，Quantum Transportation 的雲部署現在提供了處理複雜量子數據所需的可擴展基礎設施。</p>
<p>這一成就使 Quantum Transportation 能夠與量子硬件設計合作伙伴合作，並進入下一個階段：在不同架構的物理量子硬件上直接測試其無代碼解碼器。通過解決噪聲量子設備中的錯誤校正挑戰，這項技術可能在鐵路異常檢測、預測性維護和自主操作方面具有長期潛力。</p>
<p>“這次雲實施是 Quantum Transportation 的一個關鍵進展，與 Rail Vision 探索將量子人工智能創新融入運輸領域的使命完美契合，” Rail Vision 首席執行官 David BenDavid 表示。“作為 Quantum Transportation 的控股股東，我們很高興利用這個可擴展的平台來提高鐵路運營的效率和安全性，充分利用我們基於人工智能的視覺系統與量子錯誤校正之間的協同效應。”</p>
<p>Rail Vision 於 2026 年 1 月 14 日通過股份交換交易完成了對 Quantum Transportation 51% 股份的收購。Quantum Transportation 持有來自特拉維夫大學技術轉移公司 Ramot 的一項創新待申請專利的鐵路技術的獨佔子許可。</p>
<p><strong>關於 Quantum Transportation</strong></p>
<p>Quantum Transportation 提議開發一個由專利變壓器基礎的通用解碼器（PD）驅動的量子錯誤校正模擬器。該解碼器利用深度學習技術，能夠跨量子代碼進行泛化，從噪聲模式中學習，並提供可擴展且與硬件無關的錯誤校正方法。專利的深度量子錯誤校正變壓器（DQECCT）引入了一種新型機器學習解碼器，使用基於變壓器的架構預測和修正量子錯誤，結合來自奇偶校驗矩陣的掩蔽層，並優化邏輯錯誤率（LER）、比特錯誤率（BER）和噪聲估計誤差的組合損失函數。這項技術旨在在準確性和速度上超越經典解碼器（例如 MWPM），並獨特地處理故障測量場景。它適用於各種代碼，包括表面碼、顏色碼、自行車碼和乘積碼。</p>
<p><strong>關於 Rail Vision Ltd.</strong></p>
<p>Rail Vision 是一家專注於轉變鐵路安全和鐵路數據市場的早期商業化技術公司。該公司開發了尖端的、基於人工智能的行業領先技術，專門為鐵路設計。該公司開發的鐵路檢測系統旨在拯救生命、提高效率，並大幅降低鐵路運營商的費用。Rail Vision 相信，其技術將顯著提高全球鐵路安全，同時為依賴鐵路生態系統的每個人創造顯著的利益並增加價值：從使用火車進行運輸的乘客到利用鐵路運輸貨物和服務的公司。此外，該公司相信，其技術有潛力將自主列車的革命性概念變為現實。有關更多信息，請訪問 https://www.railvision.io/</p>
<p><strong>前瞻性聲明</strong></p>
<p>本新聞稿包含《私人證券訴訟改革法案》和其他證券法所定義的 “前瞻性聲明”。諸如 “預計”、“預期”、“打算”、“計劃”、“相信”、“尋求”、“估計” 等詞語及其類似表達或變體旨在識別前瞻性聲明。這些期望、信念和預測是以善意表達的。例如，Rail Vision 在討論 Quantum Transportation 與量子硬件設計合作伙伴合作並進入項目的下一個階段時，使用了前瞻性聲明，Rail Vision 的使命是探索量子人工智能創新在交通運輸領域的整合，並利用 Quantum Transportation 的可擴展平台提升鐵路運營的效率和安全性，利用其基於人工智能的視覺系統與量子糾錯之間的協同效應。然而，無法保證管理層的期望、信念和預測將會實現，實際結果可能與前瞻性聲明中所表達或指示的內容有實質性差異。前瞻性聲明受到風險和不確定性的影響，這可能導致實際表現或結果與前瞻性聲明中所表達的內容有實質性差異。有關影響公司的風險和不確定性的更詳細描述，請參考公司不時向證券交易委員會（“SEC”）提交的報告，包括但不限於在 2025 年 3 月 31 日向 SEC 提交的截至 2024 年 12 月 31 日的財政年度的 20-F 表格年報中詳細列出的風險。前瞻性聲明僅在聲明發布之日有效。公司不承擔更新前瞻性聲明以反映實際結果、後續事件或情況、假設變化或影響前瞻性信息的其他因素變化的義務，除非適用的證券法要求。如果公司更新一項或多項前瞻性聲明，不應推斷公司將就此或其他前瞻性聲明進行額外更新。提供的參考和鏈接僅為方便，網站上包含的信息並未被納入本新聞稿的引用。Rail Vision 對第三方網站的內容不承擔責任。</p>
<p><strong>聯繫方式</strong></p>
<p>David BenDavid</p>
<p>首席執行官</p>
<p>Rail Vision Ltd.</p>
<p>15 Ha'Tidhar St</p>
<p>以色列拉阿納納，郵政編碼 4366517</p>
<p>電話：+972- 9-957-7706</p>
<p><strong>投資者關係：</strong></p>
<p>Michal Efraty</p>
<p>investors@railvision.io</p>
<p><img src="https://imageproxy.pbkrs.com/https://www.globenewswire.com/newsroom/ti/query-bmY9T1RZMk1EQXpPU00zTkRVeE5ETTNJelV3TURFek1EY3dOQT09?x-oss-process=image/auto-orient,1/interlace,1/resize,w_1440,h_1440/quality,q_95/format,jpg" alt="" original-src="https://imageproxy.pbkrs.com/https://www.globenewswire.com/newsroom/ti/query-bmY9T1RZMk1EQXpPU00zTkRVeE5ETTNJelV3TURFek1EY3dOQT09"/></p>
<p><img src="https://imageproxy.pbkrs.com/https://ml.globenewswire.com/media/NWE1MjIyZmMtZTM2Yi00MDMwLWIyMGYtYmFhYmM4YjZiNDY4LTUwMDEzMDcwNC0yMDI2LTAyLTI0LWVu/tiny/Rail-Vision-Ltd-.png?x-oss-process=image/auto-orient,1/interlace,1/resize,w_1440,h_1440/quality,q_95/format,jpg" alt="" original-src="https://imageproxy.pbkrs.com/https://ml.globenewswire.com/media/NWE1MjIyZmMtZTM2Yi00MDMwLWIyMGYtYmFhYmM4YjZiNDY4LTUwMDEzMDcwNC0yMDI2LTAyLTI0LWVu/tiny/Rail-Vision-Ltd-.png"/></p>

RVSN

RVSNW

<p>Rail Vision Ltd.（納斯達克代碼：RVSN）宣佈其子公司 Quantum Transportation 已成功在 AWS 雲上實施基於變壓器的神經解碼器，這標誌着在交通運輸領域實現真實世界量子應用的重要一步。這一成就使得與量子硬件合作伙伴的協作以及在物理量子硬件上的測試成為可能。該技術旨在通過先進的錯誤糾正方法提高鐵路安全性和效率。Rail Vision 於 2026 年 1 月收購了 Quantum Transportation 51% 的股份，進一步推動其將量子人工智能創新整合到鐵路行業的使命</p>

<p><em>Quantum Transportation’s unique decoder is making another step forward by being validated in high-performance cloud environments</em></p>
<div class="lb-trans"><p>量子運輸公司的獨特解碼器在高性能雲環境中取得了新的進展</p>
</div><p>Ra’anana, Israel, Feb. 24, 2026 (GLOBE NEWSWIRE) -- Rail Vision Ltd. (Nasdaq: RVSN) (“Rail Vision” or the “Company”), an early commercialization stage technology company focused on transforming railway safety and the rail data markets, announced today that its majority owned subsidiary Quantum Transportation Ltd. (“Quantum Transportation”), a quantum computing innovator, has successfully implemented its transformer-based neural decoder on the AWS cloud, marking a significant milestone toward real-world quantum applications within the transportation sector.</p>
<div class="lb-trans"><p>以色列拉阿納納，2026 年 2 月 24 日（全球新聞通訊社）——Rail Vision Ltd.（納斯達克：RVSN）（“Rail Vision” 或 “公司”），一家專注於轉變鐵路安全和鐵路數據市場的早期商業化技術公司，今天宣佈其控股子公司 Quantum Transportation Ltd.（“Quantum Transportation”），一家量子計算創新公司，已成功在 AWS 雲上實施其基於變壓器的神經解碼器，這標誌着在運輸領域實現現實世界量子應用的重要里程碑。</p>
</div><p>Building on the recent unveiling of its transformer neural decoder, which outperformed classical quantum error correction (QEC) algorithms in simulations, and the delivery of its first prototype for universal error correction, Quantum Transportation's cloud deployment now provides the scalable infrastructure needed to process complex quantum data efficiently.</p>
<div class="lb-trans"><p>在最近推出的變壓器神經解碼器的基礎上，該解碼器在模擬中超越了經典量子錯誤校正（QEC）算法，並交付了其第一個通用錯誤校正原型，Quantum Transportation 的雲部署現在提供了處理複雜量子數據所需的可擴展基礎設施。</p>
</div><p>This achievement positions Quantum Transportation with the ability to collaborate with quantum hardware design partners and enter the next phase: direct testing of its code-agnostic decoder on physical quantum hardware across diverse architectures. By addressing error correction challenges in noisy quantum devices, this technology may have long-term potential for railway anomaly detection, predictive maintenance, and autonomous operations.</p>
<div class="lb-trans"><p>這一成就使 Quantum Transportation 能夠與量子硬件設計合作伙伴合作，並進入下一個階段：在不同架構的物理量子硬件上直接測試其無代碼解碼器。通過解決噪聲量子設備中的錯誤校正挑戰，這項技術可能在鐵路異常檢測、預測性維護和自主操作方面具有長期潛力。</p>
</div><p>“This cloud implementation is a pivotal advancement for Quantum Transportation and aligns seamlessly with Rail Vision’s mission to explore the integration of quantum-AI innovations into the transportation sector,” said David BenDavid, CEO of Rail Vision. “As Quantum Transportation’s majority owner, we are excited to leverage this scalable platform to enhance efficiency and safety in railway operations, capitalizing on the synergies between our AI-driven vision systems and quantum error correction.”</p>
<div class="lb-trans"><p>“這次雲實施是 Quantum Transportation 的一個關鍵進展，與 Rail Vision 探索將量子人工智能創新融入運輸領域的使命完美契合，” Rail Vision 首席執行官 David BenDavid 表示。“作為 Quantum Transportation 的控股股東，我們很高興利用這個可擴展的平台來提高鐵路運營的效率和安全性，充分利用我們基於人工智能的視覺系統與量子錯誤校正之間的協同效應。”</p>
</div><p>Rail Vision completed its acquisition of a 51% stake in Quantum Transportation on January 14, 2026, through a share exchange transaction. Quantum Transportation holds an exclusive sub-license for rail technologies from an innovative pending patent in quantum error correction owned by Ramot, the technology transfer company of Tel Aviv University.</p>
<div class="lb-trans"><p>Rail Vision 於 2026 年 1 月 14 日通過股份交換交易完成了對 Quantum Transportation 51% 股份的收購。Quantum Transportation 持有來自特拉維夫大學技術轉移公司 Ramot 的一項創新待申請專利的鐵路技術的獨佔子許可。</p>
</div><p><strong>About Quantum Transportation</strong></p>
<div class="lb-trans"><p><strong>關於 Quantum Transportation</strong></p>
</div><p>Quantum Transportation proposes to develop a Quantum Error Correction Simulator powered by a patented Transformer-based Universal Decoder (PD). This decoder, leveraging deep learning techniques, generalizes across quantum codes, learns from noise patterns, and delivers a scalable and hardware-agnostic approach to error correction. The patented Deep Quantum Error Correction Transformer (DQECCT) introduces a novel machine-learning decoder that predicts and refines quantum errors using transformer-based architectures, incorporates masking layers derived from parity-check matrices and optimizes a combined loss function over Logical Error Rate (LER), Bit Error Rate (BER), and Noise Estimation Error. This technology aspires to outperform classical decoders (e.g., MWPM) in both accuracy and speed and uniquely handles faulty measurement scenarios. It is adaptable to various codes - including Surface, Color, Bicycle, and Product Codes.</p>
<div class="lb-trans"><p>Quantum Transportation 提議開發一個由專利變壓器基礎的通用解碼器（PD）驅動的量子錯誤校正模擬器。該解碼器利用深度學習技術，能夠跨量子代碼進行泛化，從噪聲模式中學習，並提供可擴展且與硬件無關的錯誤校正方法。專利的深度量子錯誤校正變壓器（DQECCT）引入了一種新型機器學習解碼器，使用基於變壓器的架構預測和修正量子錯誤，結合來自奇偶校驗矩陣的掩蔽層，並優化邏輯錯誤率（LER）、比特錯誤率（BER）和噪聲估計誤差的組合損失函數。這項技術旨在在準確性和速度上超越經典解碼器（例如 MWPM），並獨特地處理故障測量場景。它適用於各種代碼，包括表面碼、顏色碼、自行車碼和乘積碼。</p>
</div><p><strong>About Rail Vision Ltd.</strong></p>
<div class="lb-trans"><p><strong>關於 Rail Vision Ltd.</strong></p>
</div><p>Rail Vision is an early commercialization stage technology company focused on transforming railway safety and the rail data markets. The company has developed cutting edge, artificial intelligence based, industry-leading technology specifically designed for railways. The company has developed its railway detection systems to save lives, increase efficiency, and dramatically reduce expenses for the railway operators. Rail Vision believes that its technology will significantly increase railway safety around the world, while creating significant benefits and adding value to everyone who relies on the train ecosystem: from passengers using trains for transportation to companies that use railways to deliver goods and services. In addition, the company believes that its technology has the potential to advance the revolutionary concept of autonomous trains into a practical reality. For more information, please visit https://www.railvision.io/</p>
<div class="lb-trans"><p>Rail Vision 是一家專注於轉變鐵路安全和鐵路數據市場的早期商業化技術公司。該公司開發了尖端的、基於人工智能的行業領先技術，專門為鐵路設計。該公司開發的鐵路檢測系統旨在拯救生命、提高效率，並大幅降低鐵路運營商的費用。Rail Vision 相信，其技術將顯著提高全球鐵路安全，同時為依賴鐵路生態系統的每個人創造顯著的利益並增加價值：從使用火車進行運輸的乘客到利用鐵路運輸貨物和服務的公司。此外，該公司相信，其技術有潛力將自主列車的革命性概念變為現實。有關更多信息，請訪問 https://www.railvision.io/</p>
</div><p><strong>Forward-Looking Statements</strong></p>
<div class="lb-trans"><p><strong>前瞻性聲明</strong></p>
</div><p>This press release contains “forward-looking statements” within the meaning of the Private Securities Litigation Reform Act and other securities laws. Words such as “expects,” “anticipates,” “intends,” “plans,” “believes,” “seeks,” “estimates” and similar expressions or variations of such words are intended to identify forward-looking statements. Such expectations, beliefs and projections are expressed in good faith. For example, Rail Vision is using forward-looking statements when it discusses Quantum Transportation collaborating with quantum hardware design partners and entering the next phase of the project, Rail Vision’s mission to explore the integration of quantum-AI innovations into the transportation sector and leveraging Quantum Transportation’s scalable platform to enhance efficiency and safety in railway operations, capitalizing on the synergies between its AI-driven vision systems and quantum error correction. However, there can be no assurance that management’s expectations, beliefs and projections will be achieved, and actual results may differ materially from what is expressed in or indicated by the forward-looking statements. Forward-looking statements are subject to risks and uncertainties that could cause actual performance or results to differ materially from those expressed in the forward-looking statements. For a more detailed description of the risks and uncertainties affecting the Company, reference is made to the Company’s reports filed from time to time with the Securities and Exchange Commission (“SEC”), including, but not limited to, the risks detailed in the Company’s annual report on Form 20-F for the fiscal year ended December 31, 2024, filed with the SEC on March 31, 2025. Forward-looking statements speak only as of the date the statements are made. The Company assumes no obligation to update forward-looking statements to reflect actual results, subsequent events or circumstances, changes in assumptions or changes in other factors affecting forward-looking information except to the extent required by applicable securities laws. If the Company does update one or more forward-looking statements, no inference should be drawn that the Company will make additional updates with respect thereto or with respect to other forward-looking statements. References and links to websites have been provided as a convenience, and the information contained on such websites is not incorporated by reference into this press release. Rail Vision is not responsible for the contents of third-party websites.</p>
<div class="lb-trans"><p>本新聞稿包含《私人證券訴訟改革法案》和其他證券法所定義的 “前瞻性聲明”。諸如 “預計”、“預期”、“打算”、“計劃”、“相信”、“尋求”、“估計” 等詞語及其類似表達或變體旨在識別前瞻性聲明。這些期望、信念和預測是以善意表達的。例如，Rail Vision 在討論 Quantum Transportation 與量子硬件設計合作伙伴合作並進入項目的下一個階段時，使用了前瞻性聲明，Rail Vision 的使命是探索量子人工智能創新在交通運輸領域的整合，並利用 Quantum Transportation 的可擴展平台提升鐵路運營的效率和安全性，利用其基於人工智能的視覺系統與量子糾錯之間的協同效應。然而，無法保證管理層的期望、信念和預測將會實現，實際結果可能與前瞻性聲明中所表達或指示的內容有實質性差異。前瞻性聲明受到風險和不確定性的影響，這可能導致實際表現或結果與前瞻性聲明中所表達的內容有實質性差異。有關影響公司的風險和不確定性的更詳細描述，請參考公司不時向證券交易委員會（“SEC”）提交的報告，包括但不限於在 2025 年 3 月 31 日向 SEC 提交的截至 2024 年 12 月 31 日的財政年度的 20-F 表格年報中詳細列出的風險。前瞻性聲明僅在聲明發布之日有效。公司不承擔更新前瞻性聲明以反映實際結果、後續事件或情況、假設變化或影響前瞻性信息的其他因素變化的義務，除非適用的證券法要求。如果公司更新一項或多項前瞻性聲明，不應推斷公司將就此或其他前瞻性聲明進行額外更新。提供的參考和鏈接僅為方便，網站上包含的信息並未被納入本新聞稿的引用。Rail Vision 對第三方網站的內容不承擔責任。</p>
</div><p><strong>Contacts</strong></p>
<div class="lb-trans"><p><strong>聯繫方式</strong></p>
</div><p>David BenDavid</p>
<div class="lb-trans"><p>David BenDavid</p>
</div><p>Chief Executive Officer</p>
<div class="lb-trans"><p>首席執行官</p>
</div><p>Rail Vision Ltd.</p>
<div class="lb-trans"><p>Rail Vision Ltd.</p>
</div><p>15 Ha'Tidhar St</p>
<div class="lb-trans"><p>15 Ha'Tidhar St</p>
</div><p>Ra'anana, 4366517 Israel</p>
<div class="lb-trans"><p>以色列拉阿納納，郵政編碼 4366517</p>
</div><p>Telephone: +972- 9-957-7706</p>
<div class="lb-trans"><p>電話：+972- 9-957-7706</p>
</div><p><strong>Investor Relations:</strong></p>
<div class="lb-trans"><p><strong>投資者關係：</strong></p>
</div><p>Michal Efraty</p>
<div class="lb-trans"><p>Michal Efraty</p>
</div><p>investors@railvision.io</p>
<div class="lb-trans"><p>investors@railvision.io</p>
</div><p><img src="https://imageproxy.pbkrs.com/https://www.globenewswire.com/newsroom/ti/query-bmY9T1RZMk1EQXpPU00zTkRVeE5ETTNJelV3TURFek1EY3dOQT09" alt="" original-src="https://www.globenewswire.com/newsroom/ti?nf=OTY2MDAzOSM3NDUxNDM3IzUwMDEzMDcwNA=="/></p>
<p><img src="https://imageproxy.pbkrs.com/https://ml.globenewswire.com/media/NWE1MjIyZmMtZTM2Yi00MDMwLWIyMGYtYmFhYmM4YjZiNDY4LTUwMDEzMDcwNC0yMDI2LTAyLTI0LWVu/tiny/Rail-Vision-Ltd-.png" alt="" original-src="https://ml.globenewswire.com/media/NWE1MjIyZmMtZTM2Yi00MDMwLWIyMGYtYmFhYmM4YjZiNDY4LTUwMDEzMDcwNC0yMDI2LTAyLTI0LWVu/tiny/Rail-Vision-Ltd-.png"/></p>

Rail Vision：量子運輸正朝着量子硬件集成的方向發展 | RVSN 股票新聞