International Association
for Cryptologic Research

There exists a mismatch between the theory and practice of cryptography in the presence of leakage. On the theoretical front, the bounded leakage model, where the adversary learns bounded-length but noiseless information about secret components, and the random probing model, where the adversary learns some internal values of a leaking implementation with some probability, are convenient abstractions to analyze the security of numerous designs. On the practical front, side-channel attacks produce long transcripts which are inherently noisy but provide information about all internal computations, and this noisiness is usually evaluated with closely related metrics like the mutual information or statistical distance. Ideally, we would like to claim that resilience to bounded leakage or random probing implies resilience to noisy leakage evaluated according to these metrics. However, prior work (Duc, Dziembowski and Faust, Eurocrypt 2014; Brian et al., Eurocrypt 2021) has shown that proving such reductions with useful parameters is challenging. In this work, we study noisy leakage models stemming from hockey-stick divergences, which generalize statistical distance and are also the basis of differential privacy. First, we show that resilience to bounded leakage and random probing implies resilience to our new noisy leakage model with improved parameters compared to models based on the statistical distance or mutual information. Second, we establish composition theorems for our model, showing that these connections extend to a setting where multiple leakages are obtained from a leaking implementation. We complement our theoretical results with a discussion of practical relevance, highlighting that (i) the reduction to bounded leakage applies to realistic leakage functions with noise levels that are decreased by several orders of magnitude compared to Brian et al., and (ii) the reduction to random probing usefully generalizes the seminal work of Duc, Dziembowski, and Faust, although it remains limited when the field size in which masking operates grows (i.e., hockey-stick divergences can better hide the field size dependency of the noise requirements, but do not annihilate it).

<jats:title>Abstract</jats:title><jats:p>We put forward two natural generalizations of predicate encryption (PE), dubbed <jats:italic>multi-key</jats:italic> and <jats:italic>multi-input</jats:italic> PE. More in details, our contributions are threefold.<jats:list list-type="bullet"> <jats:list-item> <jats:p><jats:bold>Definitions.</jats:bold> We formalize security of multi-key PE and multi-input PE following the standard indistinguishability paradigm, and modeling security both against malicious senders (i.e., corruption of encryption keys) and malicious receivers (i.e., collusions).</jats:p> </jats:list-item> <jats:list-item> <jats:p><jats:bold>Constructions.</jats:bold> We construct adaptively secure multi-key and multi-input PE supporting the conjunction of poly-many arbitrary single-input predicates, assuming the sub-exponential hardness of the learning with errors (LWE) problem.</jats:p> </jats:list-item> <jats:list-item> <jats:p><jats:bold>Applications.</jats:bold> We show that multi-key and multi-input PE for expressive enough predicates suffices for interesting cryptographic applications, including non-interactive multi-party computation (NI-MPC) and matchmaking encryption (ME).</jats:p> </jats:list-item> </jats:list> In particular, plugging in our constructions of multi-key and multi-input PE, under the sub-exponential LWE assumption, we obtain the first ME supporting <jats:italic>arbitrary policies</jats:italic> with unbounded collusions, as well as robust (resp. non-robust) NI-MPC for so-called <jats:italic>all-or-nothing</jats:italic> functions satisfying a non-trivial notion of reusability and supporting a constant (resp. polynomial) number of parties. Prior to our work, both of these applications required much heavier tools such as indistinguishability obfuscation or compact functional encryption.</jats:p>

Evolving secret sharing (Komargodski, Naor, and Yogev, TCC’16) generalizes the notion of secret sharing to the setting of evolving access structures, in which the share holders are added to the system in an online manner, and where the dealer does not know neither the access structure nor the maximum number of parties in advance. Here, the main difficulty is to distribute shares to the new players without updating the shares of old players; moreover, one would like to minimize the share size as a function of the number of players. In this paper, we initiate a systematic study of evolving secret sharing in the computational setting, where the maximum number of parties is polynomial in the security parameter, but the dealer still does not know this value, neither it knows the access structure in advance. Moreover, the privacy guarantee only holds against computationally bounded adversaries corrupting an unauthorized subset of the players. Our main result is that for many interesting, and practically relevant, evolving access structures (including graphs access structures, DNF and CNF formulas access structures, monotone circuits access structures, and threshold access structures), under standard hardness assumptions, there exist efficient secret sharing schemes with computational privacy and in which the shares are succinct (i.e., much smaller compared to the size of a natural computational representation of the evolving access structure).

Password-Authenticated Key Exchange (PAKE) allows two parties to establish a common high-entropy secret from a possibly low-entropy pre-shared secret such as a password. In this work, we provide the first PAKE protocol with subversion resilience in the framework of universal composability (UC), where the latter roughly means that UC security still holds even if one of the two parties is malicious and the honest party's code has been subverted (in an undetectable manner). We achieve this result by sanitizing the PAKE protocol from oblivious transfer (OT) due to Canetti et al. (PKC'12) via cryptographic reverse firewalls in the UC framework (Chakraborty et al., EUROCRYPT'22). This requires new techniques, which help us uncover new cryptographic primitives with sanitation-friendly properties along the way (such as OT, dual-mode cryptosystems, and signature schemes). As an additional contribution, we delve deeper in the backbone of communication required in the subversion-resilient UC framework, extending it to the {\em unauthenticated} setting, in line with the work of Barak et al. (CRYPTO'05).

We put forward two natural generalizations of predicate encryption (PE), dubbed multi-key and multi-input PE. More in details, our contributions are threefold. – Definitions. We formalize security of multi-key PE and multi-input PE following the standard indistinguishability paradigm, and modeling security both against malicious senders (i.e., corruption of encryption keys) and malicious receivers (i.e., collusions). – Constructions. We construct adaptively secure multi-key and multi-input PE supporting the conjunction of poly-many arbitrary single-input predicates, assuming the sub-exponential hardness of the learning with errors (LWE) problem. – Applications. We show that multi-key and multi-input PE for expressive enough predicates suffices for interesting cryptographic applications, including non-interactive multi-party computation (NI-MPC) and matchmaking encryption (ME). In particular, plugging in our constructions of multi-key and multi-input PE, under the sub-exponential LWE assumption, we obtain the first ME supporting arbitrary policies with unbounded collusions, as well as robust (resp. non-robust) NI-MPC for so-called all-or-nothing functions satisfying a non-trivial notion of reusability and supporting a constant (resp. polynomial) number of parties. Prior to our work, both of these applications required much heavier tools such as indistinguishability obfuscation or compact functional encryption.

Registered encryption (Garg et al., TCC'18) is an emerging paradigm that tackles the key-escrow problem associated with identity-based encryption by replacing the private-key generator with a much weaker entity known as the key curator. The key curator holds no secret information, and is responsible to: (i) update the master public key whenever a new user registers its own public key to the system; (ii) provide helper decryption keys to the users already registered in the system, in order to still enable them to decrypt after new users join the system. For practical purposes, tasks (i) and (ii) need to be efficient, in the sense that the size of the public parameters, of the master public key, and of the helper decryption keys, as well as the running times for key generation and user registration, and the number of updates, must be small. In this paper, we generalize the notion of registered encryption to the setting of functional encryption (FE). As our main contribution, we show an efficient construction of registered FE for the special case of (attribute hiding) inner-product predicates, built over asymmetric bilinear groups of prime order. Our scheme supports a large attribute universe and is proven secure in the bilinear generic group model. We also implement our scheme and experimentally demonstrate the efficiency requirements of the registered settings. Our second contribution is a feasibility result where we build registered FE for P/poly based on indistinguishability obfuscation and somewhere statistically binding hash functions.

Subversion attacks undermine security of cryptographic protocols by replacing a legitimate honest party's implementation with one that leaks information in an undetectable manner. An important limitation of all currently known techniques for designing cryptographic protocols with security against subversion attacks is that they do not automatically guarantee security in the realistic setting where a protocol session may run concurrently with other protocols. We remedy this situation by providing a foundation of reverse firewalls (Mironov and Stephens-Davidowitz, EUROCRYPT'15) in the universal composability (UC) framework (Canetti, FOCS'01 and J. ACM'20). More in details, our contributions are threefold: - We generalize the UC framework to the setting where each party consists of a core (which has secret inputs and is in charge of generating protocol messages) and a firewall (which has no secrets and sanitizes the outgoing/incoming communication from/to the core). Both the core and the firewall can be subject to different flavors of corruption, modeling different kinds of subversion attacks. For instance, we capture the setting where a subverted core looks like the honest core to any efficient test, yet it may leak secret information via covert channels (which we call specious subversion). - We show how to sanitize UC commitments and UC coin tossing against specious subversion, under the DDH assumption. - We show how to sanitize the classical GMW compiler (Goldreich, Micali and Wigderson, STOC 1987) for turning MPC with security in the presence of semi-honest adversaries into MPC with security in the presence of malicious adversaries. This yields a completeness theorem for maliciously secure MPC in the presence of specious subversion. Additionally, all our sanitized protocols are transparent, in the sense that communicating with a sanitized core looks indistinguishable from communicating with an honest core. Thanks to the composition theorem, our methodology allows, for the first time, to design subversion-resilient protocols by sanitizing different sub-components in a modular way.

We construct non-malleable codes in the split-state model with codeword length m + 3λ or m + 5λ, where m is the message size and λ is the security parameter, depending on how conservative one is. Our scheme is very simple and involves a single call to a block cipher meeting a new security notion which we dub entropic fixed-related-key security, which essentially means that the block cipher behaves like a pseudorandom permutation when queried upon inputs sampled from a distribution with sufficient min-entropy, even under related-key attacks with respect to an arbitrary but fixed key relation. Importantly, indistinguishability only holds with respect to the original secret key (and not with respect to the tampered secret key).In a previous work, Fehr, Karpman, and Mennink (ToSC 2018) used a related assumption (where the block cipher inputs can be chosen by the adversary, and where indistinguishability holds even with respect to the tampered key) to construct a nonmalleable code in the split-state model with codeword length m + 2λ. Unfortunately, no block cipher (even an ideal one) satisfies their assumption when the tampering function is allowed to be cipher-dependent. In contrast, we are able to show that entropic fixed-related-key security holds in the ideal cipher model with respect to a large class of cipher-dependent tampering attacks (including those which break the assumption of Fehr, Karpman, and Mennink).

Non-malleable codes (Dziembowski, Pietrzak and Wichs, ICS 2010 & JACM 2018) allow protecting arbitrary cryptographic primitives against related-key attacks (RKAs). Even when using codes that are guaranteed to be non-malleable against a single tampering attempt, one obtains RKA security against poly-many tampering attacks at the price of assuming perfect memory erasures. In contrast, continuously non-malleable codes (Faust, Mukherjee, Nielsen and Venturi, TCC 2014) do not suffer from this limitation, as the non-malleability guarantee holds against poly-many tampering attempts. Unfortunately, there are only a handful of constructions of continuously non-malleable codes, while standard non-malleable codes are known for a large variety of tampering families including, e.g., NC0 and decision-tree tampering, AC0, and recently even bounded polynomial-depth tampering. We change this state of affairs by providing the first constructions of continuously non-malleable codes in the following natural settings: – Against decision-tree tampering, where, in each tampering attempt, every bit of the tampered codeword can be set arbitrarily after adaptively reading up to d locations within the input codeword. Our scheme is in the plain model, can be instantiated assuming the existence of one-way functions, and tolerates tampering by decision trees of depth d = O(n1/8), where n is the length of the codeword. Notably, this class includes NC0. – Against bounded polynomial-depth tampering, where in each tampering attempt the adversary can select any tampering function that can be computed by a circuit of bounded polynomial depth (and unbounded polynomial size). Our scheme is in the common reference string model, and can be instantiated assuming the existence of time-lock puzzles and simulation-extractable (succinct) non-interactive zero-knowledge proofs.

We show that noisy leakage can be simulated in the information-theoretic setting using a single query of bounded leakage, up to a small statistical simulation error and a slight loss in the leakage parameter. The latter holds true in particular for one of the most used noisy-leakage models, where the noisiness is measured using the conditional average min-entropy (Naor and Segev, CRYPTO'09 and SICOMP'12). Our reductions between noisy and bounded leakage are achieved in two steps. First, we put forward a new leakage model (dubbed the dense leakage model) and prove that dense leakage can be simulated in the information-theoretic setting using a single query of bounded leakage, up to small statistical distance. Second, we show that the most common noisy-leakage models fall within the class of dense leakage, with good parameters. We also provide a complete picture of the relationships between different noisy-leakage models, and prove lower bounds showing that our reductions are nearly optimal. Our result finds applications to leakage-resilient cryptography, where we are often able to lift security in the presence of bounded leakage to security in the presence of noisy leakage, both in the information-theoretic and in the computational setting. Additionally, we show how to use lower bounds in communication complexity to prove that bounded-collusion protocols (Kumar, Meka, and Sahai, FOCS'19) for certain functions do not only require long transcripts, but also necessarily need to reveal enough information about the inputs.

We study non-malleable secret sharing against joint leakage and joint tampering attacks. Our main result is the first threshold secret sharing scheme in the plain model achieving resilience to noisy-leakage and continuous tampering. The above holds under (necessary) minimal computational assumptions (i.e., the existence of one-to-one one-way functions), and in a model where the adversary commits to a fixed partition of all the shares into non-overlapping subsets of at most t - 1 shares (where t is the reconstruction threshold), and subsequently jointly leaks from and tampers with the shares within each partition. We also study the capacity (i.e., the maximum achievable asymptotic information rate) of continuously non-malleable secret sharing against joint continuous tampering attacks. In particular, we prove that whenever the attacker can tamper jointly with k > t/2 shares, the capacity is at most t - k. The rate of our construction matches this upper bound. An important corollary of our results is the first non-malleable secret sharing scheme against independent tampering attacks breaking the rate-one barrier (under the same computational assumptions as above).

We introduce a new form of encryption that we name matchmaking encryption (ME). Using ME, sender S and receiver R (each with its own attributes) can both specify policies the other party must satisfy in order for the message to be revealed. The main security guarantee is that of privacy-preserving policy matching: During decryption, nothing is leaked beyond the fact that a match occurred/did not occur. ME opens up new ways of secretly communicating and enables several new applications where both participants can specify fine-grained access policies to encrypted data. For instance, in social matchmaking, S can encrypt a file containing his/her personal details and specify a policy so that the file can be decrypted only by his/her ideal partner. On the other end, a receiver R will be able to decrypt the file only if S corresponds to his/her ideal partner defined through a policy. On the theoretical side, we define security for ME, as well as provide generic frameworks for constructing ME from functional encryption. These constructions need to face the technical challenge of simultaneously checking the policies chosen by S and R, to avoid any leakage. On the practical side, we construct an efficient identity-based scheme for equality policies, with provable security in the random oracle model under the standard BDH assumption. We implement and evaluate our scheme and provide experimental evidence that our construction is practical. We also apply identity-based ME to a concrete use case, in particular for creating an anonymous bulletin board over a Tor network.

Secret sharing enables a dealer to split a secret into a set of shares, in such a way that certain authorized subsets of share holders can reconstruct the secret, whereas all unauthorized subsets cannot. Non-malleable secret sharing (Goyal and Kumar, STOC 2018) additionally requires that, even if the shares have been tampered with, the reconstructed secret is either the original or a completely unrelated one. In this work, we construct non-malleable secret sharing tolerating $p$-time {\em joint-tampering} attacks in the plain model (in the computational setting), where the latter means that, for any $p>0$ fixed {\em a priori}, the attacker can tamper with the same target secret sharing up to $p$ times. In particular, assuming one-to-one one-way functions, we obtain: - A secret sharing scheme for threshold access structures which tolerates joint $p$-time tampering with subsets of the shares of maximal size ({\em i.e.}, matching the privacy threshold of the scheme). This holds in a model where the attacker commits to a partition of the shares into non-overlapping subsets, and keeps tampering jointly with the shares within such a partition (so-called {\em selective partitioning}). - A secret sharing scheme for general access structures which tolerates joint $p$-time tampering with subsets of the shares of size $O(\sqrt{\log n})$, where $n$ is the number of parties. This holds in a stronger model where the attacker is allowed to adaptively change the partition within each tampering query, under the restriction that once a subset of the shares has been tampered with jointly, that subset is always either tampered jointly or not modified by other tampering queries (so-called {\em semi-adaptive partitioning}). At the heart of our result for selective partitioning lies a new technique showing that every one-time {\em statistically} non-malleable secret sharing against joint tampering is in fact {\em leakage-resilient} non-malleable ({\em i.e.},\ the attacker can leak jointly from the shares prior to tampering). We believe this may be of independent interest, and in fact we show it implies lower bounds on the share size and randomness complexity of statistically non-malleable secret sharing against {\em independent} tampering.

One approach toward basing public-key encryption (PKE) schemes on weak and credible assumptions is to build “stronger” or more general schemes generically from “weaker” or more restricted ones. One particular line of work in this context was initiated by Myers and Shelat (FOCS ’09) and continued by Hohenberger, Lewko, and Waters (Eurocrypt ’12), who provide constructions of multi-bit CCA-secure PKE from single-bit CCA-secure PKE. It is well known that encrypting each bit of a plaintext string independently is not CCA-secure—the resulting scheme is malleable . We therefore investigate whether this malleability can be dealt with using the conceptually simple approach of applying a suitable non-malleable code (Dziembowski et al., ICS ’10) to the plaintext and subsequently encrypting the resulting codeword bit by bit. We find that an attacker’s ability to ask multiple decryption queries requires that the underlying code be continuously non-malleable (Faust et al., TCC ’14). Since, as we show, this flavor of non-malleability can only be achieved if the code is allowed to “self-destruct,” the resulting scheme inherits this property and therefore only achieves a weaker variant of CCA security. We formalize this new notion of so-called indistinguishability under self-destruct attacks (IND-SDA) as CCA security with the restriction that the decryption oracle stops working once the attacker submits an invalid ciphertext. We first show that the above approach based on non-malleable codes yields a solution to the problem of domain extension for IND-SDA-secure PKE, provided that the underlying code is continuously non-malleable against (a reduced form of) bit-wise tampering. Then, we prove that the code of Dziembowski et al. is actually already continuously non-malleable against bit-wise tampering. We further investigate the notion of security under self-destruct attacks and combine IND-SDA security with non-malleability under chosen-ciphertext attacks (NM-CPA) to obtain the strictly stronger notion of non-malleability under self-destruct attacks (NM-SDA) . We show that NM-SDA security can be obtained from basic IND-CPA security by means of a black-box construction based on the seminal work by Choi et al. (TCC ’08). Finally, we provide a domain extension technique for building a multi-bit NM-SDA scheme from a single-bit NM-SDA scheme. To achieve this goal, we define and construct a novel type of continuous non-malleable code, called secret-state NMC , since, as we show, standard continuous NMCs are insufficient for the natural “encode-then-encrypt-bit-by-bit” approach to work.

Non-malleable codes (Dziembowski et al., ICS’10 and J. ACM’18) are a natural relaxation of error correcting/detecting codes with useful applications in cryptography. Informally, a code is non-malleable if an adversary trying to tamper with an encoding of a message can only leave it unchanged or modify it to the encoding of an unrelated value. This paper introduces continuous non-malleability, a generalization of standard non-malleability where the adversary is allowed to tamper continuously with the same encoding. This is in contrast to the standard definition of non-malleable codes, where the adversary can only tamper a single time. The only restriction is that after the first invalid codeword is ever generated, a special self-destruct mechanism is triggered and no further tampering is allowed; this restriction can easily be shown to be necessary. We focus on the split-state model, where an encoding consists of two parts and the tampering functions can be arbitrary as long as they act independently on each part. Our main contributions are outlined below. We show that continuous non-malleability in the split-state model is impossible without relying on computational assumptions. We construct a computationally secure split-state code satisfying continuous non-malleability in the common reference string (CRS) model. Our scheme can be instantiated assuming the existence of collision-resistant hash functions and (doubly enhanced) trapdoor permutations, but we also give concrete instantiations based on standard number-theoretic assumptions. We revisit the application of non-malleable codes to protecting arbitrary cryptographic primitives against related-key attacks. Previous applications of non-malleable codes in this setting required perfect erasures and the adversary to be restricted in memory. We show that continuously non-malleable codes allow to avoid these restrictions.

We revisit the concept of non-malleable secret sharing (Goyal and Kumar, STOC 2018) in the computational setting. In particular, under the assumption of one-to-one one-way functions, we exhibit a computationally private, threshold secret sharing scheme satisfying all of the following properties. Continuous non-malleability: No computationally-bounded adversary tampering independently with all the shares can produce mauled shares that reconstruct to a value related to the original secret. This holds even in case the adversary can tamper continuously, for an unbounded polynomial number of times, with the same target secret sharing, where the next sequence of tampering functions, as well as the subset of shares used for reconstruction, can be chosen adaptively based on the outcome of previous reconstructions.Resilience to noisy leakage: Non-malleability holds even if the adversary can additionally leak information independently from all the shares. There is no bound on the length of leaked information, as long as the overall leakage does not decrease the min-entropy of each share by too much.Improved rate: The information rate of our final scheme, defined as the ratio between the size of the message and the maximal size of a share, asymptotically approaches 1 when the message length goes to infinity. Previous constructions achieved information-theoretic security, sometimes even for arbitrary access structures, at the price of at least one of the following limitations: (i) Non-malleability only holds against one-time tampering attacks; (ii) Non-malleability holds against a bounded number of tampering attacks, but both the choice of the tampering functions and of the sets used for reconstruction is non-adaptive; (iii) Information rate asymptotically approaching zero; (iv) No security guarantee in the presence of leakage.

We introduce a new form of encryption that we name matchmaking encryption (ME). Using ME, sender S and receiver R (each with its own attributes) can both specify policies the other party must satisfy in order for the message to be revealed. The main security guarantee is that of privacy-preserving policy matching: During decryption nothing is leaked beyond the fact that a match occurred/did not occur.ME opens up new ways of secretly communicating, and enables several new applications where both participants can specify fine-grained access policies to encrypted data. For instance, in social matchmaking, S can encrypt a file containing his/her personal details and specify a policy so that the file can be decrypted only by his/her ideal partner. On the other end, a receiver R will be able to decrypt the file only if S corresponds to his/her ideal partner defined through a policy.On the theoretical side, we define security for ME, as well as provide generic frameworks for constructing ME from functional encryption.These constructions need to face the technical challenge of simultaneously checking the policies chosen by S and R, to avoid any leakage.On the practical side, we construct an efficient identity-based scheme for equality policies, with provable security in the random oracle model under the standard BDH assumption. We implement and evaluate our scheme and provide experimental evidence that our construction is practical. We also apply identity-based ME to a concrete use case, in particular for creating an anonymous bulletin board over a Tor network.

We show how to construct maliciously secure oblivious transfer (M-OT) from a strengthening of key agreement (KA) which we call strongly uniform KA (SU-KA), where the latter roughly means that the messages sent by one party are computationally close to uniform, even if the other party is malicious. Our transformation is black-box, almost round preserving (adding only a constant overhead of up to two rounds), and achieves standard simulation-based security in the plain model.As we show, 2-round SU-KA can be realized from cryptographic assumptions such as low-noise LPN, high-noise LWE, Subset Sum, DDH, CDH and RSA—all with polynomial hardness—thus yielding a black-box construction of fully-simulatable, round-optimal, M-OT from the same set of assumptions (some of which were not known before).

We study leakage-resilient continuously non-malleable secret sharing, as recently introduced by Faonio and Venturi (CRYPTO 2019). In this setting, an attacker can continuously tamper and leak from a target secret sharing of some message, with the goal of producing a modified set of shares that reconstructs to a message related to the originally shared value. Our contributions are two fold. In the plain model, assuming one-to-one one-way functions, we show how to obtain noisy-leakage-resilient continuous non-malleability for arbitrary access structures, in case the attacker can continuously leak from and tamper with all of the shares independently.In the common reference string model, we show how to obtain a new flavor of security which we dub bounded-leakage-resilient continuous non-malleability under selective $$k$$-partitioning. In this model, the attacker is allowed to partition the target $$n$$ shares into any number of non-overlapping blocks of maximal size $$k$$, and then can continuously leak from and tamper with the shares within each block jointly. Our construction works for arbitrary access structures, and assuming (doubly enhanced) trapdoor permutations and collision-resistant hash functions, we achieve a concrete instantiation for $$k\in O(\log n)$$. Prior to our work, there was no secret sharing scheme achieving continuous non-malleability against joint tampering, and the only known scheme for independent tampering was tailored to threshold access structures.

At ICS 2010, Dziembowski, Pietrzak and Wichs introduced the notion of non-malleable codes, a weaker form of error-correcting codes guaranteeing that the decoding of a tampered codeword either corresponds to the original message or to an unrelated value. The last few years established non-malleable codes as one of the recently invented cryptographic primitives with the highest impact and potential, with very challenging open problems and applications.In this work, we focus on so-called continuously non-malleable codes in the split-state model, as proposed by Faust et al. (TCC 2014), where a codeword is made of two shares and an adaptive adversary makes a polynomial number of attempts in order to tamper the target codeword, where each attempt is allowed to modify the two shares independently (yet arbitrarily). Achieving continuous non-malleability in the split-state model has been so far very hard. Indeed, the only known constructions require strong setup assumptions (i.e., the existence of a common reference string) and strong complexity-theoretic assumptions (i.e., the existence of non-interactive zero-knowledge proofs and collision-resistant hash functions).As our main result, we construct a continuously non-malleable code in the split-state model without setup assumptions, requiring only one-to-one one-way functions (i.e., essentially optimal computational assumptions). Our result introduces several new ideas that make progress towards understanding continuous non-malleability, and shows interesting connections with protocol-design and proof-approach techniques used in other contexts (e.g., look-ahead simulation in zero-knowledge proofs, non-malleable commitments, and leakage resilience).