NIST Round 4 & Signature Onramp
After standardizing the first PQC algorithms, NIST continued evaluating additional candidates. The fourth round initially involved four KEMs — BIKE, Classic McEliece, HQC, and SIKE — but SIKE dropped out of serious contention after a cryptanalytic break, so the round mainly evaluated BIKE, Classic McEliece, and HQC, and in 2025 selected HQC for standardization; a separate "signature onramp" call sought non-lattice signatures. On 11 March 2025, NIST selected HQC for standardization as an additional KEM alongside ML-KEM, adding diversity in its underlying code-based assumptions — not yet a published FIPS standard.
Why a fourth round?
The headline standards — ML-KEM, ML-DSA, FN-DSA — are all lattice-based. Concentrating the entire ecosystem on one mathematical family is a strategic risk: a single major breakthrough in lattice cryptanalysis could threaten all of them at once. NIST therefore pursued defense in diversity, seeking standardized algorithms built on different hard problems so that a fallback exists.
The Round 4 KEMs
| Candidate | Family | Hard problem | Notable trait | Outcome |
|---|---|---|---|---|
| BIKE | Code-based | QC-MDPC decoding | Small keys; rare decoding failures | Not selected (2025) |
| Classic McEliece | Code-based | Goppa-code decoding | Tiny ciphertext, huge public key; very conservative | Not selected in Round 4 (NIST IR 8545); may be pursued via other avenues |
| HQC | Code-based | Quasi-cyclic code decoding | Clean security analysis; selected as 2nd KEM for standardization | Selected for standardization (Mar 2025) |
All three are code-based — their security comes from the difficulty of decoding general linear error-correcting codes, a problem studied since the 1970s (the original McEliece scheme dates to 1978). This long pedigree makes them attractive as a hedge against lattice surprises.
HQC: selected as an additional KEM for standardization
On 11 March 2025, NIST announced it would standardize HQC (Hamming Quasi-Cyclic) as an additional KEM alongside ML-KEM — making it the second KEM selected for standardization and, in NIST's own press framing, the fifth PQC algorithm selected by NIST for standardization overall — though it is a KEM, not a signature algorithm. The selection rationale is documented in NIST IR 8545 (Status Report on the Fourth Round, final, Mar 2025). HQC's FIPS number has not yet been officially confirmed by NIST (it is widely expected to be FIPS 207); NIST plans to publish a draft for public comment first, with the final standard expected around 2027. The rationale was explicitly defense in diversity:
- HQC rests on code-based assumptions, mathematically distinct from ML-KEM's Module-LWE.
- It has a clean, well-understood security reduction and reasonable key/ciphertext sizes (larger than ML-KEM, but far smaller than Classic McEliece's public key).
- It gives organizations a standardized non-lattice option for hybrid deployments or as a contingency.
HQC is intended as a backup, not a replacement: ML-KEM remains the recommended primary KEM due to its smaller sizes and better performance. HQC is for those who want lattice-independent assurance.
The signature onramp
Because the signature standards are also lattice-heavy (ML-DSA and FN-DSA), NIST issued a separate call for additional signatures in 2023 — the "signature onramp." Its explicit goal was to attract non-lattice signature schemes and short signatures, broadening the portfolio beyond hash-based (SLH-DSA) and lattice options.
- Many candidates were submitted across diverse families: code-based, multivariate, isogeny-based, symmetric/MPC-in-the-head, and others.
- NIST IR 8528 (final, 2024) reported on the first round of this on-ramp.
- On 24 October 2024, NIST advanced 14 candidates to Round 2: CROSS, FAEST, HAWK, LESS, MAYO, Mirath, MQOM, PERK, QR-UOV, RYDE, SDitH, SNOVA, SQIsign, and UOV. The 6th NIST PQC Standardization Conference was held 24–26 September 2025.
- On 14 May 2026, NIST advanced 9 candidates to Round 3: FAEST, HAWK, MAYO, MQOM, QR-UOV, SDitH, SNOVA, SQIsign, and UOV — of which only HAWK is lattice-based. A roughly two-year evaluation follows, with a 7th NIST PQC Standardization Conference planned for 2027.
- The process is still under evaluation; several early submissions were broken, underscoring the value of public scrutiny.
- Any winners would complement, not replace, ML-DSA and SLH-DSA.
Standards & references
- HQC announced as a fourth-round selection (11 Mar 2025)
- NIST IR 8545 — Status Report on the Fourth Round (final, Mar 2025)
- NIST IR 8528 — Status Report on the First Round of the Additional Signatures On-Ramp (final, 2024)
- Additional Digital Signatures — Round 2 (14 candidates, 24 Oct 2024)
- Nine candidates advance to the third round of additional digital signatures (14 May 2026)
- NIST Post-Quantum Cryptography project (HQC standardization status)
NIST 第四轮与签名补充征集
在标准化首批后量子算法之后 NIST 继续评估更多候选。第四轮最初涉及 BIKE、Classic McEliece、HQC 与 SIKE;SIKE 后因密码分析突破退出实际竞争,最终主要评估 BIKE、Classic McEliece 与 HQC,并于 2025 年选择 HQC 标准化;另有一项签名补充征集寻求非格签名。2025 年 3 月 11 日,NIST 选择 HQC 进入标准化流程,作为 ML-KEM 之外的额外 KEM 标准化对象,提供编码类数学假设上的多样性补充;HQC 尚未发布为最终 FIPS 标准。
为何需要第四轮
主力标准 ML-KEM ML-DSA FN-DSA 都基于格。将整个生态系统押注于单一数学家族是一种战略风险针对格的密码分析一旦取得重大突破便可能同时危及这几种算法。因此 NIST 推行多样性防御另行寻求建立在不同困难问题之上的标准化算法以保留退路。
第四轮的 KEM
| 候选 | 家族 | 困难问题 | 显著特征 | 结果 |
|---|---|---|---|---|
| BIKE | 编码类 | QC-MDPC 译码 | 密钥小偶发译码失败 | 未选定 2025 |
| Classic McEliece | 编码类 | Goppa 码译码 | 密文极小公钥巨大极为保守 | 第四轮未入选(NIST IR 8545)可能通过其他渠道推进 |
| HQC | 编码类 | 准循环码译码 | 安全分析清晰 被选为第二个进入标准化的 KEM | 选定 2025 年 3 月 |
三者皆为编码类其安全性源自一般线性纠错码译码的困难性这一问题自 1970 年代起即被研究最初的 McEliece 方案可追溯至 1978 年。这一悠久渊源使它们成为对冲格类算法风险的有力之选。
HQC 已入选标准化的追加 KEM
2025 年 3 月 11 日 NIST 宣布将把 HQC 汉明准循环标准化为 ML-KEM 之外的额外 KEM使其成为第二个进入标准化流程的 KEM;按 NIST 官方新闻表述,HQC 将成为第五个被 NIST 选定用于标准化的 PQC 算法,但它是 KEM,不是签名算法。选定依据记录于 NIST IR 8545 第四轮状态报告(正式,2025 年 3 月)。HQC 的 FIPS 编号尚待 NIST 正式确认(外界普遍预期为 FIPS 207);NIST 计划先发布草案并征求意见,最终标准预计在 2027 年左右完成。其理由明确为多样性防御
- HQC 建立在编码类假设上数学上与 ML-KEM 的模 LWE 截然不同。
- 它具备清晰且广为理解的安全归约以及合理的密钥/密文尺寸大于 ML-KEM 但远小于 Classic McEliece 的公钥。
- 它为组织提供了一个标准化的非格选项用于混合部署或作为应急方案。
HQC 定位为备份而非替代因尺寸更小性能更优 ML-KEM 仍是推荐的主力 KEM。HQC 适合那些希望获得不依赖格类假设之安全保证的场景。
签名补充征集
由于签名标准同样偏重格 ML-DSA 与 FN-DSANIST 于 2023 年另行发起额外签名征集即签名补充征集。其明确目标是吸引非格签名方案与短签名在哈希类 SLH-DSA 与格选项之外拓宽组合。
- 来自不同家族的众多方案应征提交编码类多变量同源以及对称/MPC-in-the-head 等。
- NIST IR 8528 正式 2024 报告了本次补充征集第一轮的结果。
- 2024 年 10 月 24 日 NIST 将 14 个候选推进至第二轮CROSS FAEST HAWK LESS MAYO Mirath MQOM PERK QR-UOV RYDE SDitH SNOVA SQIsign 与 UOV。第六届 NIST PQC 标准化大会于 2025 年 9 月 24 至 26 日举行。
- 2026 年 5 月 14 日 NIST 将 9 个候选推进至第三轮FAEST HAWK MAYO MQOM QR-UOV SDitH SNOVA SQIsign 与 UOV其中仅 HAWK 基于格。随后将进行约两年的评估第七届 NIST PQC 标准化大会预计于 2027 年举行。
- 该过程仍在评估中若干早期提交已被攻破凸显了公开审查的价值。
- 任何胜出者都将是 ML-DSA 与 SLH-DSA 的补充而非替代。
标准与参考
- HQC 被宣布为第四轮选定方案 2025 年 3 月 11 日
- NIST IR 8545 第四轮状态报告 正式 2025 年 3 月
- NIST IR 8528 额外签名第一轮状态报告 正式 2024
- 额外数字签名第二轮 14 个候选 2024 年 10 月 24 日
- 九个候选推进至额外数字签名第三轮 2026 年 5 月 14 日
- NIST 后量子密码项目 HQC 标准化进展