Submissions claiming to resolve a grand challenge, such as the P vs. NP problem, may be rejected without further consideration.
Submissions containing an essential part that seems to be generated by AI tools and/or are written in a way that humans will find hard to understand, may be rejected regardless of their merits. ECCC is intended for communication among humans.
For more details see the Call for Papers.In the last few days, a Denial of Service attack was launched on universities in Israel, leading the administrators of the Israel Academic network to block access to it from the global internet. Consequently, websites such as ECCC have been accessible only from within the Israeli and European academic networks.
It seems that this blocking was just removed, and we hope it will not be put back in the future.
Needless to say, deciding on such blocking is not in our control, but we do apologize for this disruption of service.
Matrix concentration for Markov chains was initiated in the expander-walk setting by Garg, Lee, Song, and Srivastava'18 [GLSS18]. However, the constant obtained in [GLSS18] is quite loose, and it is natural to ask whether a tighter proof can yield the same constant as in the independent matrix concentration setting. In ... more >>>
We study interactive proofs of proximity (IPP) for model evaluation: a resource-limited verifier interacts with an untrusted prover, typically the model owner, to certify statistical properties of a model under an unknown input distribution. Our formulation is shaped by the constraints of practical evaluation: it separates sampling the input distribution ... more >>>
A subspace design is a collection of subspaces $H_1,\ldots,H_n$ of $\mathbb{F}_q^k$ with the property that no low-dimensional subspace $W$ intersects the collection ``too much’’. Subspace designs and related objects in linear-algebraic pseudorandomness have found a broad range of applications, ranging from list decoding and recovery, to derandomizing algorithms.
We ... more >>>
