Cryptographic research has long wrestled with a persistent efficiency barrier in standard prime-order groups: while advanced cryptographic protocols allow users to commit to large vectors of data and prove individual elements within them efficiently, those proofs have stubbornly refused to shrink below a logarithmic size. Now, a new foundational study by researchers at ETH Zurich sheds light on this limitation, proving that this long-standing barrier is not a failure of imagination among cryptographers, but an inherent mathematical reality.
In a newly released paper titled “Bulletproofs are Optimal: Lower Bounds for Vector Commitments from Fiat-Shamir in Pairing-Free Groups,” researchers Marian Dietz and Emanuele Giunta from ETH Zurich establish rigorous lower bounds for vector commitments and related primitives in pairing-free prime-order groups. Their findings demonstrate that the logarithmic scaling seen in celebrated protocols like Bulletproofs cannot be fundamentally surpassed in the plain discrete logarithm setting without making heavy compromises elsewhere.
The research brings formal mathematical closure to a central design question that has challenged the zero-knowledge proof and cryptographic commitment communities for over a decade. By analyzing the fundamental limits of interactive public-coin membership proofs within Maurer’s Generic Group Model, the authors show that state-of-the-art schemes achieving logarithmic proof sizes are, in a strict asymptotic sense, optimal.
Understanding Vector Commitments and the Logarithmic Barrier
To grasp the significance of the ETH Zurich team’s findings, one must understand the role that vector commitments play in modern cryptography. Vector commitments allow multiple parties or a single prover to commit to an ordered vector of data items in a concise manner—often resulting in a single small commitment string. Later, the prover can open the commitment at any desired position, providing a succinct cryptographic proof that a specific value resides at a specific index within the committed vector, without needing to reveal the rest of the underlying data.
In plain, known-order, and pairing-free prime-order groups, all state-of-the-art constructions rely on a tried-and-true recipe: the combination of a generalized Pedersen commitment scheme with an inner product argument. This design lineage includes foundational cryptographic contributions such as the celebrated constructions by Bootle et al., introduced at EUROCRYPT 2016, and Bünz et al.’s Bulletproofs framework, presented at the IEEE Symposium on Security and Privacy (S&P) in 2018, alongside various subsequent improvements and optimizations.
These widely utilized protocols achieve opening proofs consisting of $Theta(log n)$ group elements, where $n$ represents the length or size of the committed vector. Beyond their compact proofs, these constructions provide a rich suite of highly useful cryptographic properties, including malleability, subvector opening capabilities, and a transparent setup that requires no trusted setup ceremony.
Despite these overwhelming advantages, breaking the $O(log n)$ efficiency barrier in the plain discrete logarithm setting has historically proven to be an extraordinarily difficult problem. Cryptographers striving to construct constant-size vector commitments—where proof sizes remain static regardless of how large the vector grows—found themselves repeatedly hitting a wall in pairing-free groups.
This frustrating limitation stood in stark contrast with alternative cryptographic settings. For example, in cryptographic environments utilizing pairing-friendly groups or groups of unknown order, constructions featuring constant-size commitments and constant-size opening proofs have been known and successfully deployed for over a decade. The persistent gap between pairing-friendly settings and pairing-free prime-order groups left many wondering whether clever new mathematical techniques might eventually break the logarithmic barrier in the plain discrete logarithm setting, or if the setting itself imposed an unavoidable penalty.
Investigating Inherent Limitations in the Generic Group Model
To resolve this open question, Marian Dietz and Emanuele Giunta set out to investigate whether the limitation of logarithmic proofs is an inherent mathematical property of constructions based on prime-order groups, rather than merely an artifact of current protocol designs.
Rather than focusing solely on vector commitments in isolation, the researchers broadened their analytical scope by examining accumulators. An accumulator is a weaker cryptographic primitive than a vector commitment, meaning that proving lower bounds for accumulators makes the resulting theorem significantly more general and robust. If an inherent limitation applies to accumulators, it naturally extends to any stronger primitive built upon them, including full vector commitments.
Using Maurer’s Generic Group Model, the authors analyzed interactive public-coin membership proofs. Their methodological approach allowed them to rigorously track how provers and verifiers interact within generic groups, stripping away any algebraic artifacts specific to particular curve implementations and focusing purely on the structural limits of generic group operations.
The core implication of their newly proven lower bound is profound for protocol designers. The research establishes that in the plain discrete logarithm setting, when utilizing pairing-free prime-order groups, at least one of three undesirable conditions must inevitably occur for any accumulator or vector commitment scheme.
First, the commitment itself must have a super-constant size, meaning it cannot remain a compact, fixed-size string as the vector scales. Second, the opening proof must contain an omega of log $n$ ($Omega(log n)$) number of group elements, validating the asymptotic optimality of existing logarithmic schemes. Third, if one attempts to bypass these constraints, the opening proof must stretch to a length of $n^1-o(1)$, which completely undermines the succinctness required for efficient verification.
Broader Implications for Functional Commitments and Inner Product Arguments
The theoretical weight of the new paper extends far beyond standard vector commitments. Dietz and Giunta demonstrate that their newly derived lower bounds naturally apply to a wider family of cryptographic primitives that directly imply accumulators and vector commitments.
This broad extension encompasses polynomial commitments and functional commitments, which are crucial backbone components for modern transparent zero-knowledge succinct non-interactive arguments of knowledge (zk-SNARKs). Furthermore, the bounds capture inner product arguments, which serve as the mathematical engine driving Bulletproofs and many related zero-knowledge proof systems.
By proving that these foundational building blocks are bounded by the same asymptotic constraints when instantiated in pairing-free prime-order groups, the research provides a definitive map of the cryptographic landscape. Protocol designers working on privacy-preserving technologies, decentralized systems, and verifiable computation can now view the performance characteristics of pairing-free algebraic tools through a definitive lens. Rather than expending research capital attempting to achieve constant-size openings in the plain discrete logarithm setting via standard techniques, the cryptographic community can use these lower bounds to guide the optimal deployment and engineering of existing logarithmic frameworks.
The paper, formally cataloged as Paper 2026/2868 on the Cryptology ePrint Archive, offers comprehensive mathematical proofs and formalizations that solidify the theoretical foundations of pairing-free cryptography. As the cryptographic community continues to advance scalable, transparent, and trustless protocols, the work of Dietz and Giunta ensures that researchers build upon a clear and mathematically verified understanding of what is fundamentally achievable in standard prime-order groups.






