Obfuscated reversible circuits can do CCA2 encryption, and even leak into FE/HE territory if you squint hard enough.

Adaptor signatures, now for BLS too, so the usual atomic-swap and ZK-contingent-payment tricks can stop special-casing Schnorr.

HE CNN inference paper: channel-interleaved packing and a small rotation-key set, trimming the usual encrypted-conv tax.

Paper on self-tuning ring-oscillator TRNGs and long-term stability, with NIST tests, because entropy still needs babysitting.

New dual LWE attack variant mixes modulus switching with lossy source coding, shaving FFT costs and nudging Kyber bounds.

APN permutations over GR(4,2) get classified, and the paper drags differential uniformity beyond prime fields into ring land.

Anonymous comms on constant-degree expanders, via onion routing and random walks, match complete-network asymptotics.

Sublinear encrypted comparisons for PNN search, by clustering and homomorphic PQ, because fully homomorphic pain needed pruning.

Can TFHE bootstrapping get cheaper by splitting lookup tables and doing homomorphic div/rem?

Trace-only segmentation of cipher implementations, no labels or metadata, just repetition-scale estimates and stable cores.

Weak-key distinguishers pop out of a split-and-cancel oracle, extending integral attacks to SIMON, SPECK, PRESENT, and GIFT.

Polynomial-time key recovery breaks Facto-DSA, with all proposed parameter sets falling in under a minute on a laptop.

ePrint preprint on lookup-based zkML costs, showing activation structure won't cut per-lookup proving work, because tables win.

Paper on constant-round secure comparison for Shamir MPC, with private decision-tree eval and lower latency, because trees needed help.

Boolean functions get partitioned by inner products, then analyzed for slice complexity, GSW-style homomorphic eval, and crypto metrics.

Hybrid CPU-GPU sumcheck backend for Spartan provers, keeps polynomial state on GPU until it stops paying rent.

Can a Halo2 zkVM prove synthesized hardware is power-side-channel safe without leaking the netlist? Apparently, yes.

Rejection sampling, the old side-channel workhorse, beats CDT here with fewer masked ANDs for lattice Gaussians (incl. HAWK).

A bit-operation cost model for XL plus Wiedemann/Berlekamp-Massey, with formulas and experiments for MQ challenge and NIST candidates.

Can you fine-tune a model privately with off-the-shelf MPC/HE and still benchmark it on a real PPML task? PriFT says yes.

Walsh LUTs on lazy CKKS bits speed up AES transciphering, because apparently XORs deserved a detour.

Proves no ideal arithmetic correlation for several ternary/quaternary/6-ary sequences under stated algebraic conditions.

AB-IPFE gets a two-level hierarchy, plus pairing-based adaptive-secure constructions for expressive access control.

Can you verify a blind watermark without leaking the params? TIM says yes, using Nova+Spartan for the ZK plumbing.

Paper proving black-box PIR with preprocessing still hits lower bounds, so the free lunch is, naturally, imaginary.

Can TALUS's threshold ML-DSA hide the key? This note says Feldman commitments and signature broadcasts leak it anyway.

So Cramer’s rule is just Lagrange interpolation in disguise, and this note spells out the MPC/secret-sharing angle?

A preprint saying the HAWK “Guessing Game” attack seems to run into class-number hell, not the claimed polynomial time.

PSOs at PSI speed, by swapping curve tricks for RLWE OKVS decoding and batching the whole mess.

Adds k-idempotent bent functions in the partial spread class; existence for all k|n up to 3, and k=1,2 are fully characterized.