Paper on Groth16-based private cross-chain verification for Substrate, with Circom/SnarkJS and XCM sims, because of course.

Can conjugating a block cipher really cut multiplications? This paper says yes, with SKINNY getting leaner.

Pipelined FPGA post-processing for QKD, trading logic for throughput and resource efficiency like crypto hardware loves to do.

SVP/BDD time-vs-success tradeoffs, then LWE/SIS reductions that care about time-bounded adversaries, not just PPT ones.

Logarithmic signatures for multi-signer stats, because linear homomorphic signatures apparently weren't niche enough.

A field-choice taxonomy for SNARKs/STARKs, explaining why Plonky2/3, Stwo, and Binius ditched the usual big primes.

How little sign leakage breaks Falcon, HAWK, and ML-DSA? Apparently enough to turn key recovery into halfspace learning.

Gröbner bases beat the classical modular-polynomial setup here, for 2^k and 3^k supersingular isogenies.

SDDT models skipped bitwise ops in bitslice ciphers, then uses that to key-recover PIPO with fewer faults.

HEAD-FL: federated learning with round-adaptive DP and verifiable homomorphic aggregation, because fixed noise was too honest.

Unlinkability gets a full formal model here, plus anonymous credentials with native predicate proofs and BBS-based instantiations.

Can MPC handle weighted votes and changing committees? This preprint says yes, with malicious security and state handover.

Paper audits S/MIME's post-quantum story and introduces SMIME-PQCheck to classify mail as PQ, hybrid, classical, or unsafe.

SoK on FRI lineage, from proximity gaps to Johnson bounds, with the usual transparent SNARK security footnotes.

Two-server PIR where one box only serves public encodings and the other computes, with LPN-backed poly costs and lookup-table apps.

Can you get a robust DVRF in an async network without the usual liveness drama? This paper claims yes, via FROST and ROAST.

ePrint paper on a lattice attack against generalized cubic Pell RSA, recovering keys past the old bound. Cute.

Formalizes dead-man-switch crypto in UC, with constructions for post-mortem release or provable destruction of a lone secret.

Spain lets you prove approximate numerical computations directly, instead of pretending real numbers fit into exact constraints.

Can secret replication turn PRFs into PCFs for VOLE-style triples? This paper says yes, with symmetric crypto only.

3PaaS adds post-compromise security via a third party, with unlinkable identities, revocation, and ZK blind-signature proofs.

HPCC gives single-cycle masked 3-input multiplies and 2-cycle AES S-boxes in the PINI glitch model.

IACR preprint on P^2HDS, a hybrid classical/PQ encryption framework for IoMT edge sharing, because one cipher was too easy.

Generalizes CKKS key-switching decomposition to many inputs/keys, so batched rotations/relinearizations get less noise and less waste.

RotPIR paper on sublinear single-server PIR with FHE preprocessing via oblivious rotation, because one server was too easy.

How tight are DFR bounds for encoded lattice PKE? This paper tightens them with Barnes-Wall structure and mixed-noise chi-square models.

Is this just a GA for SVP with an outer GA tuning the inner one, or have we truly run out of ideas?

Can AEGIS be bitsliced without becoming a crime scene? This paper says yes, with CT ARM, RISC-V, wasm, and JS code.

How much cheaper do SIS attacks get when you stop hand-waving the sieve distribution? This paper says: quite a bit, including Falcon-256.

Anonymous swaps, until the threshold key gets to deanonymize you in disputes, via a ring-adaptor signature compiler with a Python artifact.