Grzegorz (Greg) Głuch

Position verification schemes are interactive protocols where entities prove their physical location to others; this enables interactive proofs for statements of the form "I am at a location L." Although secure position verification cannot be achieved with classical protocols (even with computational assumptions), they are feasible with quantum protocols. In this paper we introduce the notion of zero-knowledge position verification, which generalizes position verification in two ways: 1. enabling entities to prove more sophisticated statements about their locations at different times (for example, "I was NOT near location L at noon yesterday"). 2. maintaining privacy for any other detail about their true location besides the statement they are proving. We construct zero-knowledge position verification from standard position verification and post-quantum one-way functions. The central tool in our construction is a primitive we call position commitments, which allow entities to privately commit to their physical position in a particular moment, which is then revealed at some later time.

With the increased deployment of large language models (LLMs), one concern is their potential misuse for generating harmful content. Our work studies the alignment challenge, with a focus on filters to prevent the generation of unsafe information. Two natural points of intervention are the filtering of the input prompt before it reaches the model, and filtering the output after generation. Our main results demonstrate computational challenges in filtering both prompts and outputs. First, we show that there exist LLMs for which there are no efficient prompt filters: adversarial prompts that elicit harmful behavior can be easily constructed, which are computationally indistinguishable from benign prompts for any efficient filter. Our second main result identifies a natural setting in which output filtering is computationally intractable. All of our separation results are under cryptographic hardness assumptions. In addition to these core findings, we also formalize and study relaxed mitigation approaches, demonstrating further computational barriers. We conclude that safety cannot be achieved by designing filters external to the LLM internals (architecture and weights); in particular, black-box access to the LLM will not suffice. Based on our technical results, we argue that an aligned AI system's intelligence cannot be separated from its judgment.

We formalize and analyze the trade-off between backdoor-based watermarks and adversarial defenses, framing it as an interactive protocol between a verifier and a prover. While previous works have primarily focused on this trade-off, our analysis extends it by identifying transferable attacks as a third, counterintuitive but necessary option. Our main result shows that for all learning tasks, at least one of the three exists: a watermark, an adversarial defense, or a transferable attack. By transferable attack, we refer to an efficient algorithm that generates queries indistinguishable from the data distribution and capable of fooling all efficient defenders. Using cryptographic techniques, specifically fully homomorphic encryption, we construct a transferable attack and prove its necessity in this trade-off. Furthermore, we show that any task that satisfies our notion of a transferable attack implies a cryptographic primitive, thus requiring the underlying task to be computationally complex. Finally, we show that tasks of bounded VC-dimension allow adversarial defenses against all attackers, while a subclass allows watermarks secure against fast adversaries.

Given an n×n array of positive numbers, find a tiling using at most p rectangles minimizing the maximum rectangle weight (sum of covered entries). We prove it is NP-hard to approximate this problem within a factor of 4/3.

Nonlocality and its connections to entanglement are fundamental features of quantum mechanics that have found numerous applications in quantum information science. A set of correlations is said to be nonlocal if it cannot be reproduced by spacelike-separated parties sharing randomness and performing local operations. An important practical consideration is that the runtime of the parties has to be shorter than the time it takes light to travel between them. One way to model this restriction is to assume that the parties are computationally bounded. We therefore initiate the study of nonlocality under computational assumptions and derive the following results: (a) We define the set 𝖭𝖾𝖫 (not-efficiently-local) as consisting of all bipartite states whose correlations arising from local measurements cannot be reproduced with shared randomness and polynomial-time local operations. (b) Under the assumption that the Learning With Errors problem cannot be solved in quantum polynomial-time, we show that 𝖭𝖾𝖫=𝖤𝖭𝖳, where 𝖤𝖭𝖳 is the set of all bipartite entangled states (pure and mixed). This is in contrast to the standard notion of nonlocality where it is known that some entangled states, e.g. Werner states, are local. (c) We prove that if 𝖭𝖾𝖫=𝖤𝖭𝖳 unconditionally, then 𝙱𝚀𝙿≠𝙿𝙿. (d) Using (c), we show that a certain natural class of 1-round delegated quantum computation protocols that are sound against 𝙿𝙿 provers cannot exist.

This paper presents a surprising connection between recent advances in delegation of quantum computation to adversarial robustness. Imagine an adversary A that is in between a classifier C and a source S producing samples to be classified. Can we design an interactive protocol between C and the A, in which A would convince C that the samples being sent for classification really came from S? Using ideas from delegation of quantum computation we show it is possible to do it if A receives quantum states from S and A communicates with C classically. Why do we need quantum mechanics? Assume that S is uniform on {0,1}^n. How can A convince C that the samples he sent really came from S and were not handcrafted by S (if A received classical samples from S)? This seems to us be an almost impossible problem to solve, as we discuss in the paper.

There are two sides of this paper. The technical one and potential applications to adversarial robustness. On the technical side we show exponential separation between two models: the standard PAC-learning model and an Equivalence-Query model. Thinking of adversarial examples as counterexamples, our result can be viewed as theoretical confirmation of findings that structured adversarial examples can speed up learning. We also discuss implications for adversarial robustness beyond norm-bounded threat models.

How hard is it to adversarially attack a network with only black-box access? We analyze a spectrum of threat models indexed by query access and give lower bounds on the number of queries needed to match white-box attacks, in terms of entropy of decision boundaries. We also analyze classical learning algorithms on synthetic tasks and prove meaningful security guarantees.

Given a graph G that can be partitioned into k disjoint expanders, we construct a small-space data structure (a spectral clustering oracle) that allows quickly classifying vertices according to their cluster. Our main result gives sublinear query time with strong per-cluster guarantees, using dot-product access to a spectral embedding via estimates of short random walks and a carefully analyzed linear transformation.

Given black-box access to a high-accuracy classifier f, we show how to construct a classifier g that remains accurate and is provably robust to adversarial L2 perturbations. Our approach builds on randomized smoothing and uses geometric tools like random partitions and doubling dimension to bound adversarial error in terms of optimal error.

We show that determinacy remains undecidable even without regularity assumptions: it is undecidable for finite unions of conjunctive path queries, strengthening prior undecidability results for Regular Path Queries.

We solve a long-standing open problem by proving that Query Determinacy is undecidable for Regular Path Queries, a foundational query language for graph databases.