pith. sign in

arxiv: 2606.21799 · v1 · pith:VISR37G6new · submitted 2026-06-19 · 💻 cs.CC · cs.CR

Towards a Doubly Efficient IP=PSPACE

Liyan Chen , Matthew M. Hong , Yael Tauman Kalai , Zoe Xi This is my paper

Pith reviewed 2026-06-26 12:14 UTC · model grok-4.3

classification 💻 cs.CC cs.CR
keywords interactive proofsPSPACEdoubly efficient proofsIP=PSPACEdirect constructiontime-bounded computationinteractive proof systems
0
0 comments X

The pith

Every language in PSPACE decidable in time n to the O(log n) admits a doubly efficient interactive proof.

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The paper shows that PSPACE languages decided by a Turing machine in time T(n) equal to n to the O(log n) have interactive proof systems where the prover runs in time polynomial in T(n) and the verifier runs in time polynomial in n. This extends the prior best regime of n to the O of the square root of log n over log log n. The protocol is built directly from the underlying primitives rather than by first constructing batch interactive proofs and invoking them as a black box. A sympathetic reader would care because the result widens the known range of time bounds that admit such efficient proofs and offers a simpler route that may support further extensions.

Core claim

We show that every language in PSPACE decidable by a Turing machine in time T(n)=n^{O(log n)} admits a doubly efficient interactive proof system: the prover runs in time polynomial in T(n), and the verifier runs in time polynomial in n. This extends the best previously known regime for such proof systems from T(n)=n^{O(√{log n / log log n})}, established by Berger, Goyal, Hong, and Kalai (FOCS 2025), to T(n)=n^{O(log n)}. Beyond improving the range of T, our protocol is substantially simpler than previous doubly efficient proofs for time-bounded PSPACE. Earlier constructions proceed indirectly: they first build batch interactive proofs and then invoke them as a black box to obtain doubly eff

What carries the argument

The direct construction of the interactive proof protocol that composes the underlying primitives for time-bounded PSPACE without an intermediate batch-proof layer.

If this is right

  • The known regime of time bounds admitting doubly efficient interactive proofs for PSPACE is extended from n^{O(√{log n / log log n})} to n^{O(log n)}.
  • Doubly efficient protocols for these languages can be obtained without first building and invoking batch interactive proofs.
  • Future improvements to the time bound can proceed by refining the direct construction rather than the indirect batch-proof route.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • Iterating or strengthening the direct construction might reach polynomial time bounds and thereby establish that IP equals PSPACE with both parties efficient.
  • The same direct-composition technique could apply to other classes that have interactive proofs but lack known doubly efficient versions.
  • Avoiding the batch-proof intermediate step may reduce the overhead that previously limited how far the time bound could be pushed.

Load-bearing premise

The direct construction correctly composes the underlying interactive proof primitives for the stated time bound without hidden efficiency losses or unstated restrictions on the PSPACE machine.

What would settle it

A concrete language in PSPACE that is decidable in time n^{O(log n)} yet requires the verifier in every interactive proof to run in superpolynomial time in n.

Figures

Figures reproduced from arXiv: 2606.21799 by Liyan Chen, Matthew M. Hong, Yael Tauman Kalai, Zoe Xi.

Figure 1
Figure 1. Figure 1: Grid of protocol dimensions Batch(t, k). Moving left decreases time t, and moving down decreases batch size k. The goal is to reduce to the case where both t and k are small. When k is too tiny for instance reduction and t is still large, the prover can reduce t by λ by sending λ many equally spaced “midpoints” between the length-t path of xstart and xend for each length-t transition statement (xstart, xen… view at source ↗
Figure 2
Figure 2. Figure 2: Grid of the sub-protocol dimensions that we actually recur to, and their respective batches [PITH_FULL_IMAGE:figures/full_fig_p019_2.png] view at source ↗
read the original abstract

We show that every language in PSPACE decidable by a Turing machine in time $T(n)=n^{O(\log n)}$ admits a doubly efficient interactive proof system: the prover runs in time polynomial in T(n), and the verifier runs in time polynomial in n. This extends the best previously known regime for such proof systems from $T(n)=n^{O(\sqrt{\log n / \log\log n})}$, established by Berger, Goyal, Hong, and Kalai (FOCS 2025), to $T(n)=n^{O(\log n)}$. Beyond improving the range of T, our protocol is substantially simpler than previous doubly efficient proofs for time-bounded PSPACE. Earlier constructions proceed indirectly: they first build batch interactive proofs and then invoke them as a black box to obtain doubly efficient protocols. In contrast, we give a direct construction. This not only simplifies the proof but also points to a more promising route for future improvements.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

2 major / 1 minor

Summary. The manuscript claims that every language in PSPACE decidable by a Turing machine in time T(n)=n^{O(log n)} admits a doubly efficient interactive proof system, with prover runtime polynomial in T(n) and verifier runtime polynomial in n. This improves the prior regime of T(n)=n^{O(√{log n / log log n})} from Berger, Goyal, Hong, and Kalai (FOCS 2025). The proof is via a direct (non-black-box) construction of the protocol, which the authors argue is simpler than prior indirect approaches that first build batch interactive proofs.

Significance. If the central claim holds with the stated efficiency bounds, the result meaningfully extends the known range of time bounds for which doubly efficient IPs exist for PSPACE and demonstrates that a direct construction can achieve better parameters than black-box composition. The emphasis on simplicity could aid future work toward closing the gap to full IP=PSPACE, though the current bound remains far from polynomial time.

major comments (2)
  1. [Abstract / main theorem statement] The central claim that the direct construction yields prover time exactly poly(T(n)) for T(n)=n^{O(log n)} without hidden efficiency losses is load-bearing but unsupported by any explicit overhead analysis or scaling equations in the provided text; the abstract asserts the bounds but supplies no protocol steps, composition lemmas, or parameter tracking to confirm that polylog(T(n)) factors from the underlying primitives do not appear.
  2. [Abstract / construction overview] The manuscript states that the construction 'correctly composes the underlying interactive proof primitives' for the improved regime, yet provides no verification that the composition preserves the polynomial verifier time in n when the PSPACE machine is restricted only by the T(n)=n^{O(log n)} bound; this assumption is the weakest link identified in the skeptic note and requires a concrete lemma or calculation.
minor comments (1)
  1. [Abstract] The abstract references the prior bound from Berger et al. (FOCS 2025) but does not include a citation or comparison table; adding this would improve context.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for the thoughtful review and for identifying points where the presentation of efficiency bounds could be strengthened. We address each major comment below. The full manuscript contains the detailed construction, lemmas, and parameter tracking that support the claimed bounds; we will add explicit cross-references and a short overhead summary to the introduction to make this clearer.

read point-by-point responses

  1. Referee: [Abstract / main theorem statement] The central claim that the direct construction yields prover time exactly poly(T(n)) for T(n)=n^{O(log n)} without hidden efficiency losses is load-bearing but unsupported by any explicit overhead analysis or scaling equations in the provided text; the abstract asserts the bounds but supplies no protocol steps, composition lemmas, or parameter tracking to confirm that polylog(T(n)) factors from the underlying primitives do not appear.

    Authors: The full manuscript (Sections 3–5) provides the direct construction together with explicit overhead analysis. Lemma 4.2 tracks the precise polynomial degree and polylog factors arising from the underlying sumcheck and low-degree test primitives; Theorem 5.1 then composes them to show that the total prover runtime is O(T(n)^c) for a fixed constant c independent of the log n exponent in T. No hidden super-polynomial losses occur. We will add a one-paragraph summary of this tracking (with the relevant lemma citations) to the introduction in the revised version. revision: partial

  2. Referee: [Abstract / construction overview] The manuscript states that the construction 'correctly composes the underlying interactive proof primitives' for the improved regime, yet provides no verification that the composition preserves the polynomial verifier time in n when the PSPACE machine is restricted only by the T(n)=n^{O(log n)} bound; this assumption is the weakest link identified in the skeptic note and requires a concrete lemma or calculation.

    Authors: Section 4.3 contains the composition lemma (Lemma 4.3) that verifies the verifier runtime remains poly(n) under the stated T(n) bound. The proof proceeds by induction on the number of composition steps and shows that each step multiplies the verifier cost by only a fixed polynomial in n (independent of T), because the query complexity of the underlying primitives is polylog(T(n)) and T(n) = n^{O(log n)} keeps the product polynomial. We will include an explicit statement of this lemma in the introduction and add a short calculation illustrating the polynomial degree. revision: partial

Circularity Check

0 steps flagged

No significant circularity; new direct construction is independent of prior self-citation.

full rationale

The paper claims a new direct construction that extends the prior regime from overlapping-author work (Berger et al. FOCS 2025) to T(n)=n^{O(log n)} with prover time poly(T(n)) and verifier time poly(n). The abstract presents this as a simplification via direct (non-black-box) composition rather than invoking the prior result as a load-bearing primitive or redefining inputs in terms of outputs. No equations, fitted parameters, or self-referential definitions are indicated that would reduce the claimed bounds to the inputs by construction. The derivation chain is therefore self-contained against external benchmarks.

Axiom & Free-Parameter Ledger

0 free parameters · 1 axioms · 0 invented entities

The result rests on standard definitions and composition properties of interactive proofs and PSPACE; no free parameters, invented entities, or ad-hoc axioms are visible in the abstract.

axioms (1)
  • standard math Standard definitions and closure properties of interactive proof systems (IP) and polynomial-space computation (PSPACE)
    The claim is stated in terms of these established classes and the existence of interactive proofs for them.

pith-pipeline@v0.9.1-grok · 5701 in / 1258 out tokens · 23226 ms · 2026-06-26T12:14:08.417421+00:00 · methodology

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Reference graph

Works this paper leans on

300 extracted references · 294 canonical work pages

  1. [1]

    Minimum Disclosure Proofs of Knowledge , journal =

    Gilles Brassard and David Chaum and Claude Cr. Minimum Disclosure Proofs of Knowledge , journal =

  2. [2]

    3rd ACM STOC , pages=

    The complexity of theorem-proving procedures , author=. 3rd ACM STOC , pages=

  3. [3]

    , title =

    Schwartz, Jacob T. , title =. Journal of the ACM , volume =

  4. [4]

    Journal of the ACM , month = oct, pages =

    Shamir, Adi , title =. 1992 , publisher =. doi:10.1145/146585.146609 , journal =

    work page doi:10.1145/146585.146609 1992

  5. [5]

    EUROSAM '79: International Symposium on Symbolic and Algebraic Computation , series =

    Zippel, Richard , title =. EUROSAM '79: International Symposium on Symbolic and Algebraic Computation , series =

  6. [6]

    L. J. Comput. Syst. Sci. , volume =. 1988 , _url =. doi:10.1016/0022-0000(88)90028-1 , timestamp =

    work page doi:10.1016/0022-0000(88)90028-1 1988

  7. [7]

    and Rothblum, Ron D

    Reingold, Omer and Rothblum, Guy N. and Rothblum, Ron D. , title =. Proceedings of the 33rd Computational Complexity Conference , articleno =. 2018 , isbn =

    2018

  8. [8]

    Modern Computer Algebra , publisher=

    von zur Gathen, Joachim and Gerhard, Juergen , year=. Modern Computer Algebra , publisher=

  9. [9]

    Forgery Attacks on SipHash

    Kosuke Sasaki and Rikuto Kurahara and Kosei Sakamoto and Takanori Isobe. Forgery Attacks on SipHash. doi:10.1007/978-981-96-9095-4_1

    work page doi:10.1007/978-981-96-9095-4_1

  10. [10]

    Cryptanalysis of Fruit- F : Exploiting Key-Derivation Weaknesses and Initialization Vulnerabilities

    Subhadeep Banik and Hailun Yan. Cryptanalysis of Fruit- F : Exploiting Key-Derivation Weaknesses and Initialization Vulnerabilities. doi:10.1007/978-981-96-9095-4_2

    work page doi:10.1007/978-981-96-9095-4_2

  11. [11]

    Exploring Key-Recovery-Friendly Differential Distinguishers for SM4 and Their Performance in Differential Attacks

    Bingqing Li and Ling Sun. Exploring Key-Recovery-Friendly Differential Distinguishers for SM4 and Their Performance in Differential Attacks. doi:10.1007/978-981-96-9095-4_3

    work page doi:10.1007/978-981-96-9095-4_3

  12. [12]

    Inner Product Masked Integral Distinguishers and Integral Sets over Large Finite Fields - Applications to MiMC , CIMINION and Chaghri

    Weizhe Wang and Deng Tang and Haoyang Wang. Inner Product Masked Integral Distinguishers and Integral Sets over Large Finite Fields - Applications to MiMC , CIMINION and Chaghri. doi:10.1007/978-981-96-9095-4_4

    work page doi:10.1007/978-981-96-9095-4_4

  13. [13]

    Improved Differential Meet-in-the-Middle Cryptanalysis on SIMON and Piccolo

    Weiqing Deng and Jianing Zhang and Haoyang Wang. Improved Differential Meet-in-the-Middle Cryptanalysis on SIMON and Piccolo. doi:10.1007/978-981-96-9095-4_5

    work page doi:10.1007/978-981-96-9095-4_5

  14. [14]

    Strengthening Key Scheduling of AES -256 with Minimal Software Modifications

    Shoma Kawakami and Kazuma Taka and Atsushi Tanaka and Tatsuya Ishikawa and Takanori Isobe. Strengthening Key Scheduling of AES -256 with Minimal Software Modifications. doi:10.1007/978-981-96-9095-4_6

    work page doi:10.1007/978-981-96-9095-4_6

  15. [15]

    Ideal Transformations for Public Key Encryption

    Yao Cheng and Xianhui Lu and Ziyi Li. Ideal Transformations for Public Key Encryption. doi:10.1007/978-981-96-9095-4_7

    work page doi:10.1007/978-981-96-9095-4_7

  16. [16]

    Indifferentiability Separations in Ideal Public Key Encryption: Explicit vs

    Yao Cheng and Xianhui Lu and Ziyi Li and Yongjian Yin. Indifferentiability Separations in Ideal Public Key Encryption: Explicit vs. Implicit Rejection. doi:10.1007/978-981-96-9095-4_8

    work page doi:10.1007/978-981-96-9095-4_8

  17. [17]

    Compressed Sigma Protocols: New Model and Aggregation Techniques

    Yuxi Xue and Tianyu Zheng and Shang Gao and Bin Xiao and Man Ho Au. Compressed Sigma Protocols: New Model and Aggregation Techniques. doi:10.1007/978-981-96-9095-4_9

    work page doi:10.1007/978-981-96-9095-4_9

  18. [18]

    Glitter : A Fully Adaptive and Tightly Secure Threshold Signature

    Shaolong Tang and Peng Jiang and Liehuang Zhu. Glitter : A Fully Adaptive and Tightly Secure Threshold Signature. doi:10.1007/978-981-96-9095-4_10

    work page doi:10.1007/978-981-96-9095-4_10

  19. [19]

    Faster VOLEitH Signatures from All-But-One Vector Commitment and Half-Tree

    Dung Bui and Kelong Cong and Cyprien Delpech de Saint Guilhem. Faster VOLEitH Signatures from All-But-One Vector Commitment and Half-Tree. doi:10.1007/978-981-96-9095-4_11

    work page doi:10.1007/978-981-96-9095-4_11

  20. [20]

    Three-Round (Robust) Threshold ECDSA from Threshold CL Encryption

    Bowen Jiang and Guofeng Tang and Haiyang Xue. Three-Round (Robust) Threshold ECDSA from Threshold CL Encryption. doi:10.1007/978-981-96-9095-4_12

    work page doi:10.1007/978-981-96-9095-4_12

  21. [21]

    Lattice Attack with EHNP : Key Recovery from Two ECDSA Signatures and Breaking the Information-Theoretic Limit

    Tianyou Tang and Shuqin Fan. Lattice Attack with EHNP : Key Recovery from Two ECDSA Signatures and Breaking the Information-Theoretic Limit. doi:10.1007/978-981-96-9095-4_13

    work page doi:10.1007/978-981-96-9095-4_13

  22. [22]

    FlexiADKG : A Flexible Asynchronous Distributed Key Generation Protocol with Constant Round Complexity

    Yang Yang and Bingyu Li and Zhenyang Ding and Qianhong Wu and Bo Qin and Qin Wang. FlexiADKG : A Flexible Asynchronous Distributed Key Generation Protocol with Constant Round Complexity. doi:10.1007/978-981-96-9095-4_14

    work page doi:10.1007/978-981-96-9095-4_14

  23. [23]

    TEAKEX : TESLA -Authenticated Group Key Exchange

    Qinyi Li and Lise Millerjord and Colin Boyd. TEAKEX : TESLA -Authenticated Group Key Exchange. doi:10.1007/978-981-96-9095-4_15

    work page doi:10.1007/978-981-96-9095-4_15

  24. [24]

    Liu and Shirui Pan and Tsz Hon Yuen

    Qishuang Fu and Joseph K. Liu and Shirui Pan and Tsz Hon Yuen. SoK : A Deep Dive Into Anti-money Laundering Techniques for Blockchain Cryptocurrencies. doi:10.1007/978-981-96-9095-4_16

    work page doi:10.1007/978-981-96-9095-4_16

  25. [25]

    Advanced Temporal Graph Embedding for Detecting Fraudulent Transactions on Complex Blockchain Transactional Networks

    Jianbin Gao and Ansu Badjie and Qi Xia and Patrick Mukala and Hu Xia and Grace Mupoyi Ntuala. Advanced Temporal Graph Embedding for Detecting Fraudulent Transactions on Complex Blockchain Transactional Networks. doi:10.1007/978-981-96-9095-4_17

    work page doi:10.1007/978-981-96-9095-4_17

  26. [26]

    Walnut: A Generic Framework with Enhanced Scalability for BFT Protocols

    Lei Tian and Chenke Wang and Yu Long and Xian Xu and Mingchao Wan and Chunmiao Li and Shifeng Sun and Dawu Gu. Walnut: A Generic Framework with Enhanced Scalability for BFT Protocols. doi:10.1007/978-981-96-9095-4_18

    work page doi:10.1007/978-981-96-9095-4_18

  27. [27]

    PPSCCC : Privacy-Preserving Scalable Cross-Chain Communication Among Multiple Blockchains Based on Parent-Child Blockchain

    Hideaki Miyaji and Noriaki Kamiyama. PPSCCC : Privacy-Preserving Scalable Cross-Chain Communication Among Multiple Blockchains Based on Parent-Child Blockchain. doi:10.1007/978-981-96-9095-4_19

    work page doi:10.1007/978-981-96-9095-4_19

  28. [28]

    Towards Quantum Security of Hirose Compression Function and Romulus- H

    Shaoxuan Zhang and Chun Guo and Meiqin Wang. Towards Quantum Security of Hirose Compression Function and Romulus- H. doi:10.1007/978-981-96-9098-5_1

    work page doi:10.1007/978-981-96-9098-5_1

  29. [29]

    Efficient Multi-instance Vector Commitment and Application to Post-quantum Signatures

    Dung Bui. Efficient Multi-instance Vector Commitment and Application to Post-quantum Signatures. doi:10.1007/978-981-96-9098-5_2

    work page doi:10.1007/978-981-96-9098-5_2

  30. [30]

    Breaking the Shield: Novel Fault Attacks on CRYSTALS -Dilithium

    Dixiao Du and Yuejun Liu and Yiwen Gao and Jingdian Ming and Hao Yuan and Yongbin Zhou. Breaking the Shield: Novel Fault Attacks on CRYSTALS -Dilithium. doi:10.1007/978-981-96-9098-5_3

    work page doi:10.1007/978-981-96-9098-5_3

  31. [31]

    Efficient Revocable Identity-Based Encryption from Middle-Product LWE

    Takumi Nishimura and Atsushi Takayasu. Efficient Revocable Identity-Based Encryption from Middle-Product LWE. doi:10.1007/978-981-96-9098-5_4

    work page doi:10.1007/978-981-96-9098-5_4

  32. [32]

    Code-Based Fully Dynamic Accountable Ring Signatures and Group Signatures Using the Helper Methodology

    Rishiraj Bhattacharyya and Sreehari Kollath and Christophe Petit. Code-Based Fully Dynamic Accountable Ring Signatures and Group Signatures Using the Helper Methodology. doi:10.1007/978-981-96-9098-5_5

    work page doi:10.1007/978-981-96-9098-5_5

  33. [33]

    Partial Key Exposure Attacks on UOV and Its Variants

    Yuki Seto and Hiroki Furue and Atsushi Takayasu. Partial Key Exposure Attacks on UOV and Its Variants. doi:10.1007/978-981-96-9098-5_6

    work page doi:10.1007/978-981-96-9098-5_6

  34. [34]

    Unbounded Multi-hop Proxy Re-encryption with HRA Security: An LWE -Based Optimization

    Xiaohan Wan and Yang Wang and Haiyang Xue and Mingqiang Wang. Unbounded Multi-hop Proxy Re-encryption with HRA Security: An LWE -Based Optimization. doi:10.1007/978-981-96-9098-5_7

    work page doi:10.1007/978-981-96-9098-5_7

  35. [35]

    Fiat - Shamir with Rejection and Rotation

    Xianhui Lu and Yongjian Yin and Dingding Jia and Jingnan He and Yamin Liu and Yijian Liu and Hongbo Liu. Fiat - Shamir with Rejection and Rotation. doi:10.1007/978-981-96-9098-5_8

    work page doi:10.1007/978-981-96-9098-5_8

  36. [36]

    Amoeba: More Flexible RLWE -Based KEM

    Qingfeng Wang and Li-Ping Wang. Amoeba: More Flexible RLWE -Based KEM. doi:10.1007/978-981-96-9098-5_9

    work page doi:10.1007/978-981-96-9098-5_9

  37. [37]

    Get Rid of Templates: A Chosen-Ciphertext Attack on ML - KEM with a DPA -Based Self-comparison Oracle

    Zhenzhi Lai and Udaya Parampalli. Get Rid of Templates: A Chosen-Ciphertext Attack on ML - KEM with a DPA -Based Self-comparison Oracle. doi:10.1007/978-981-96-9098-5_10

    work page doi:10.1007/978-981-96-9098-5_10

  38. [38]

    Accountability for Server Misbehavior in Homomorphic Secret Sharing

    Xinzhou Wang and Shifeng Sun and Dawu Gu and Yuan Luo. Accountability for Server Misbehavior in Homomorphic Secret Sharing. doi:10.1007/978-981-96-9098-5_11

    work page doi:10.1007/978-981-96-9098-5_11

  39. [39]

    Jiang and Jingjing Fan and Man Ho Au and Siu Ming Yiu

    Zejiu Tan and Junping Wan and Zoe L. Jiang and Jingjing Fan and Man Ho Au and Siu Ming Yiu. High-Precision Homomorphic Modular Reduction for CKKS Bootstrapping. doi:10.1007/978-981-96-9098-5_12

    work page doi:10.1007/978-981-96-9098-5_12

  40. [40]

    Refined Error Management for Gate Bootstrapping

    Chunling Chen and Xianhui Lu and Binwu Xiang and Bowen Huang and Ruida Wang and Yijian Liu. Refined Error Management for Gate Bootstrapping. doi:10.1007/978-981-96-9098-5_13

    work page doi:10.1007/978-981-96-9098-5_13

  41. [41]

    Compact Lifting for NTT -Unfriendly Modulus

    Ying Liu and Xianhui Lu and Yu Zhang and Ruida Wang and Ziyao Liu and Kunpeng Wang. Compact Lifting for NTT -Unfriendly Modulus. doi:10.1007/978-981-96-9098-5_14

    work page doi:10.1007/978-981-96-9098-5_14

  42. [42]

    Guaranteed Termination Asynchronous Complete Secret Sharing with Lower Communication and Optimal Resilience

    Ying Cai and Chengyi Qin and Mingqiang Wang. Guaranteed Termination Asynchronous Complete Secret Sharing with Lower Communication and Optimal Resilience. doi:10.1007/978-981-96-9098-5_15

    work page doi:10.1007/978-981-96-9098-5_15

  43. [43]

    Solving Generalized Approximate Divisor Multiples Problems

    Naoki Shimoe and Noboru Kunihiro. Solving Generalized Approximate Divisor Multiples Problems. doi:10.1007/978-981-96-9098-5_16

    work page doi:10.1007/978-981-96-9098-5_16

  44. [44]

    Comparing and Improving Frequency Estimation Perturbation Mechanisms Under Local Differential Privacy

    She Sun and Jiafei Wu and Jian Yang and Li Zhou and Huiwen Wu. Comparing and Improving Frequency Estimation Perturbation Mechanisms Under Local Differential Privacy. doi:10.1007/978-981-96-9101-2_1

    work page doi:10.1007/978-981-96-9101-2_1

  45. [45]

    Strong Federated Authentication With Password-Based Credential Against Identity Server Corruption

    Changsong Jiang and Chunxiang Xu and Guomin Yang and Li Duan and Jing Wang. Strong Federated Authentication With Password-Based Credential Against Identity Server Corruption. doi:10.1007/978-981-96-9101-2_2

    work page doi:10.1007/978-981-96-9101-2_2

  46. [46]

    Anonymous Credentials with Credential Redaction and Its Application to SSI -Based Plug& Charge for Shared Vehicles

    Kyosuke Hatsugai and Kyoichi Asano and Yuki Sawai and Yohei Watanabe and Mitsugu Iwamoto. Anonymous Credentials with Credential Redaction and Its Application to SSI -Based Plug& Charge for Shared Vehicles. doi:10.1007/978-981-96-9101-2_3

    work page doi:10.1007/978-981-96-9101-2_3

  47. [47]

    Direction-Oriented Smooth Sensitivity and Its Application to Genomic Statistical Analysis

    Akito Yamamoto and Tetsuo Shibuya. Direction-Oriented Smooth Sensitivity and Its Application to Genomic Statistical Analysis. doi:10.1007/978-981-96-9101-2_4

    work page doi:10.1007/978-981-96-9101-2_4

  48. [48]

    Sentence Embedding Generation Method for Differential Privacy Protection

    Yangyang Liu and Wanqi Wang and Jingyu Hua. Sentence Embedding Generation Method for Differential Privacy Protection. doi:10.1007/978-981-96-9101-2_5

    work page doi:10.1007/978-981-96-9101-2_5

  49. [49]

    KD - IBMRKE - PPFL : A Privacy-Preserving Federated Learning Framework Integrating Knowledge Distillation and Identity-Based Multi-receiver Key Encapsulation

    Yuan Li and Changji Wang and Shiwen Hu. KD - IBMRKE - PPFL : A Privacy-Preserving Federated Learning Framework Integrating Knowledge Distillation and Identity-Based Multi-receiver Key Encapsulation. doi:10.1007/978-981-96-9101-2_6

    work page doi:10.1007/978-981-96-9101-2_6

  50. [50]

    Ebron Jr

    Sheldon C. Ebron Jr. and Meiying Zhang and Kan Yang. Identifying the Truth of Global Model: A Generic Solution to Defend Against Byzantine and Backdoor Attacks in Federated Learning. doi:10.1007/978-981-96-9101-2_7

    work page doi:10.1007/978-981-96-9101-2_7

  51. [51]

    RAGLeak : Membership Inference Attacks on RAG -Based Large Language Models

    Kaiyue Feng and Guangsheng Zhang and Huan Tian and Heng Xu and Yanjun Zhang and Tianqing Zhu and Ming Ding and Bo Liu. RAGLeak : Membership Inference Attacks on RAG -Based Large Language Models. doi:10.1007/978-981-96-9101-2_8

    work page doi:10.1007/978-981-96-9101-2_8

  52. [52]

    Zhenzhu Chen and Yansong Gao and Anmin Fu and Fanjian Zeng and Boyu Kuang and Robert H. Deng. DeGain : Detecting GAN -Based Data Inversion in Collaborative Deep Learning. doi:10.1007/978-981-96-9101-2_9

    work page doi:10.1007/978-981-96-9101-2_9

  53. [53]

    FRFL : Fair and Robust Federated Learning Incentive Model Based on Game Theory

    Haocheng Ye and Lu Zhou and Hao Wang and Chunpeng Ge. FRFL : Fair and Robust Federated Learning Incentive Model Based on Game Theory. doi:10.1007/978-981-96-9101-2_10

    work page doi:10.1007/978-981-96-9101-2_10

  54. [54]

    DPFedSub : A Differentially Private Federated Learning with Randomized Subspace Descend

    Huiwen Wu and Chuan Ma and Xueran Li and Deyi Zhang and Xiaohan Li and She Sun. DPFedSub : A Differentially Private Federated Learning with Randomized Subspace Descend. doi:10.1007/978-981-96-9101-2_11

    work page doi:10.1007/978-981-96-9101-2_11

  55. [55]

    MG -Det: Deepfake Detection with Multi-granularity

    Ahmed Asiri and Luoyu Chen and Zhiyi Tian and Xiaoyu Ding and Shui Yu. MG -Det: Deepfake Detection with Multi-granularity. doi:10.1007/978-981-96-9101-2_12

    work page doi:10.1007/978-981-96-9101-2_12

  56. [56]

    LPIA : Label Preference Inference Attack Against Federated Graph Learning

    Jiaxue Bai and Lu Shi and Yang Liu and Weizhe Zhang. LPIA : Label Preference Inference Attack Against Federated Graph Learning. doi:10.1007/978-981-96-9101-2_13

    work page doi:10.1007/978-981-96-9101-2_13

  57. [57]

    Kanhere and Jiamou Sun and Sanjay K

    Lihua Wang and Jiaojiao Jiang and Salil S. Kanhere and Jiamou Sun and Sanjay K. Jha and Zhenchang Xing. DARA : Enhancing Vulnerability Alignment via Adaptive Reconstruction and Dual-Level Attention. doi:10.1007/978-981-96-9101-2_14

    work page doi:10.1007/978-981-96-9101-2_14

  58. [58]

    Zeroth-Order Federated Private Tuning for Pretrained Large Language Models

    Xiaoyu Zhang and Yong Lin and Meixia Miao and Jian Lou and Jin Li and Xiaofeng Chen. Zeroth-Order Federated Private Tuning for Pretrained Large Language Models. doi:10.1007/978-981-96-9101-2_15

    work page doi:10.1007/978-981-96-9101-2_15

  59. [59]

    Guanqin Zhang and Feng Xu and H. M. N. Dilum Bandara and Shiping Chen and Yulei Sui. Understanding the Robustness of Machine-Unlearning Models. doi:10.1007/978-981-96-9101-2_16

    work page doi:10.1007/978-981-96-9101-2_16

  60. [60]

    Mitigating the Unprivileged User Namespaces Based Privilege Escalation Attacks with Linux Capabilities

    Jingzi Meng and Yuewu Wang and Lingguang Lei and Chunjing Kou and Peng Wang and Huawei Lu. Mitigating the Unprivileged User Namespaces Based Privilege Escalation Attacks with Linux Capabilities. doi:10.1007/978-981-96-9101-2_17

    work page doi:10.1007/978-981-96-9101-2_17

  61. [61]

    SoK : From Systematization to Best Practices in Fuzz Driver Generation

    Qian Yan and Minhuan Huang and Huayang Cao and Shuaibing Lu. SoK : From Systematization to Best Practices in Fuzz Driver Generation. doi:10.1007/978-981-96-9101-2_18

    work page doi:10.1007/978-981-96-9101-2_18

  62. [62]

    Facial Authentication Security Evaluation Against Deepfake Attacks in Mobile Apps

    Chuer Yu and Siyi Xia and Haoyu Wang and Xia Liu and Zonghui Wang and Lirong Fu and Zhiyuan Wan and Yandong Gao and Yang Xiang and Wenzhi Chen. Facial Authentication Security Evaluation Against Deepfake Attacks in Mobile Apps. doi:10.1007/978-981-96-9101-2_19

    work page doi:10.1007/978-981-96-9101-2_19

  63. [63]

    EAPIR : Efficient and Authenticated Private Information Retrieval with Fast Server Processing

    Hua Shen and Xinjie Li and Zhen Fan and Ge Wu and Mingwu Zhang. EAPIR : Efficient and Authenticated Private Information Retrieval with Fast Server Processing. doi:10.1007/978-981-96-9101-2_20

    work page doi:10.1007/978-981-96-9101-2_20

  64. [64]

    Arash Mahboubi and Hamed Aboutorab and Seyit Camtepe and Hang Thanh Bui and Khanh Luong and Keyvan Ansari and Shenlu Wang and Bazara I. A. Barry. Ransomware Encryption Detection: Adaptive File System Analysis Against Evasive Encryption Tactics. doi:10.1007/978-981-96-9101-2_21

    work page doi:10.1007/978-981-96-9101-2_21

  65. [65]

    Receiver-Initiated Updatable Public Key Encryption: Construction, Security and Application

    Jiahao Xuan. Receiver-Initiated Updatable Public Key Encryption: Construction, Security and Application. doi:10.1007/978-981-96-9101-2_22

    work page doi:10.1007/978-981-96-9101-2_22

  66. [66]

    Robust and Privacy-Preserving Dynamic Average Consensus with Individual Weight

    Yuanyuan Zhang and Yu Liu and Yahui Wang and Tianqing Zhu and Mingwu Zhang. Robust and Privacy-Preserving Dynamic Average Consensus with Individual Weight. doi:10.1007/978-981-96-9101-2_23

    work page doi:10.1007/978-981-96-9101-2_23

  67. [67]

    Improving RSA Cryptanalysis: Combining Continued Fractions and Coppersmith's Techniques

    Mengce Zheng and Yansong Feng and Abderrahmane Nitaj and Yanbin Pan. Improving RSA Cryptanalysis: Combining Continued Fractions and Coppersmith's Techniques. doi:10.1007/978-981-96-9101-2_24

    work page doi:10.1007/978-981-96-9101-2_24

  68. [68]

    Shortest Printable Shellcode Encoding Algorithm Based on Dynamic Bitwidth Selection

    Guoan Liu and Jian Lin and Weiyu Dong and Jiaan Liu and Tieming Liu. Shortest Printable Shellcode Encoding Algorithm Based on Dynamic Bitwidth Selection. doi:10.1007/978-981-96-9101-2_25

    work page doi:10.1007/978-981-96-9101-2_25

  69. [69]

    Bridging Clone Detection and Industrial Compliance: A Practical Pipeline for Enterprise Codebases

    Xiaowei Zhang and Shigang Liu and Jun Zhang and Yang Xiang. Bridging Clone Detection and Industrial Compliance: A Practical Pipeline for Enterprise Codebases. doi:10.1007/978-981-96-9101-2_26

    work page doi:10.1007/978-981-96-9101-2_26

  70. [70]

    The Offline Quantum Attack Against Modular Addition Variant of Even - Mansour Cipher

    Fangzhou Liu and Xueqi Zhu and Ruozhou Xu and Danping Shi and Peng Wang. The Offline Quantum Attack Against Modular Addition Variant of Even - Mansour Cipher. doi:10.1007/978-981-97-5025-2_1

    work page doi:10.1007/978-981-97-5025-2_1

  71. [71]

    Known-Key Attack on GIFT -64 and GIFT -64 [g_0^c] Based on Correlation Matrices

    Xiaomeng Sun and Wenying Zhang and Ren \'e Rodr \'i guez and Huimin Liu. Known-Key Attack on GIFT -64 and GIFT -64 [g_0^c] Based on Correlation Matrices. doi:10.1007/978-981-97-5025-2_2

    work page doi:10.1007/978-981-97-5025-2_2

  72. [72]

    On the Security Bounds for Block Ciphers Without Whitening Key Addition Against Integral Distinguishers

    Fanyang Zeng and Tian Tian. On the Security Bounds for Block Ciphers Without Whitening Key Addition Against Integral Distinguishers. doi:10.1007/978-981-97-5025-2_3

    work page doi:10.1007/978-981-97-5025-2_3

  73. [73]

    Tight Multi-user Security of Ascon and Its Large Key Extension

    Bishwajit Chakraborty and Chandranan Dhar and Mridul Nandi. Tight Multi-user Security of Ascon and Its Large Key Extension. doi:10.1007/978-981-97-5025-2_4

    work page doi:10.1007/978-981-97-5025-2_4

  74. [74]

    Differential Distinguishing Attacks on SNOW - V , SNOW -Vi and KCipher -2

    Rikuto Kurahara and Kosei Sakamoto and Yuto Nakano and Takanori Isobe. Differential Distinguishing Attacks on SNOW - V , SNOW -Vi and KCipher -2. doi:10.1007/978-981-97-5025-2_5

    work page doi:10.1007/978-981-97-5025-2_5

  75. [75]

    Efficient Search for Optimal Permutations of Refined Type- II Generalized Feistel Structures

    Xiaodan Li and Wenling Wu and Yuhan Zhang and Ee Duan. Efficient Search for Optimal Permutations of Refined Type- II Generalized Feistel Structures. doi:10.1007/978-981-97-5025-2_6

    work page doi:10.1007/978-981-97-5025-2_6

  76. [76]

    Man Chen and Yuyue Chen and Rui Zong and ZengPeng Li and Zoe L. Jiang. F- FHEW : High-Precision Approximate Homomorphic Encryption with Batch Bootstrapping. doi:10.1007/978-981-97-5025-2_7

    work page doi:10.1007/978-981-97-5025-2_7

  77. [77]

    Robin Jadoul and Axel Mertens and Jeongeun Park and Hilder V. L. Pereira. NTRU -Based FHE for Larger Key and Message Space. doi:10.1007/978-981-97-5025-2_8

    work page doi:10.1007/978-981-97-5025-2_8

  78. [78]

    Jiang and Jun Zhou and Junbin Fang and Zhenfu Cao

    Yi Huang and Junping Wan and Zoe L. Jiang and Jun Zhou and Junbin Fang and Zhenfu Cao. An Efficient Integer-Wise ReLU on TFHE. doi:10.1007/978-981-97-5025-2_9

    work page doi:10.1007/978-981-97-5025-2_9

  79. [79]

    HERatio : Homomorphic Encryption of Rationals Using Laurent Polynomials

    Luke Harmon and Gaetan Delavignette and Hanes Oliveira. HERatio : Homomorphic Encryption of Rationals Using Laurent Polynomials. doi:10.1007/978-981-97-5025-2_10

    work page doi:10.1007/978-981-97-5025-2_10

  80. [80]

    TFHE Bootstrapping: Faster, Smaller and Time-Space Trade-Offs

    Ruida Wang and Benqiang Wei and Zhihao Li and Xianhui Lu and Kunpeng Wang. TFHE Bootstrapping: Faster, Smaller and Time-Space Trade-Offs. doi:10.1007/978-981-97-5025-2_11

    work page doi:10.1007/978-981-97-5025-2_11

Showing first 80 references.