The paper presents an interactive probabilistically checkable proof (PCP) protocol that allows a polynomial‑time verifier to check the approximate consistency of a probabilistic predictor defined by two circuits, P and Q. By evaluating these circuits at a few points and querying a proof oracle that encodes a witnessing probability distribution, the verifier can confirm that the predictor’s many conditional‑probability claims are self‑consistent. The authors also establish that the problem of verifying l₂‑approximate consistency for explicit probabilistic claims lies in NP, with certificates of size O(mn + log B), and show how to eliminate dependence on the input bit‑precision B through a small additive gap.
By Orr Paradise, Oliver Richardson, Yoshua Bengio, Shafi Goldwasser
The paper investigates whether interactive arguments for oracle‑aided AI computations can be zero‑knowledge, meaning the verifier learns nothing beyond the correctness of the output. It proves that, in general, zero‑knowledge proofs for all oracle‑aided computations are impossible, even in the random oracle model, and this impossibility extends to debate protocols. However, if the oracle signs each answer with a cryptographic signature, then every oracle‑aided computation can be verified in zero‑knowledge with efficient provers and verifiers, assuming only collision‑resistant hash functions.
By Alessandro Chiesa, Ziyi Guan, Burcu Yildiz
arXiv:2607. 03561v1 Announce Type: new Abstract: As AI models continue to develop powerful capabilities, it becomes critical that we are able to verify that their output is aligned with our intentions.
By Liyan Chen, Yael Tauman Kalai, Zoe Xi
arXiv:2603. 25414v4 Announce Type: replace-cross Abstract: A prevailing assumption in machine learning is that model correctness must be enforced after the fact.
By Houston Haynes
arXiv:2608.31120v1 Announce Type: new
Abstract: The motivation for this paper is the investigation of the trade-offs implicit in probabilistic models used in machine learning. Models are often used t...
By Guy Emerson
arXiv:2608. 16438v1 Announce Type: new Abstract: In a world where valuable artifacts are increasingly created, completed, or processed by LLMs, the central economic question is not only what the LLM can produce, but what \emph{value} remains in the inputs (i.
By Rafael Pass
arXiv:2607. 15528v1 Announce Type: new Abstract: Following Goldwasser, Rothblum, Shafer, and Yehudayoff, who defined a framework for interactive proofs of learning [ITCS'21], we initiate the study of non-interactive proofs of learning.
By Michael Ngo, Michael P. Kim
The paper investigates the feasibility of exact truthfulness in calibration measures for sequential binary prediction. It proves that exact truthfulness cannot coexist with completeness and soundness, even when outcomes are independent. The authors then provide two reductions that transform any base calibration measure into additively or multiplicatively approximately truthful ones, achieving a multiplicative truthfulness guarantee that improves upon previous results.
By Anagha Gokul, Jason Hartline, Lunjia Hu, Jonathan Ullman, Yifan Wu
arXiv:2603. 18046v2 Announce Type: replace-cross Abstract: We present NanoZK, a zero-knowledge proof system for verifiable LLM inference: clients and third-party auditors check that a provider executed the advertised model on a committed input without learning weights or activations.
By Zhaohui Wang
arXiv:2609.25388v1 Announce Type: cross
Abstract: A classical question in statistics is which observable quantities to condition on when drawing inferences about unobservable targets. For conformal p...
By Xuelin Yang, Baihe Huang, Yilong Hou, Guido Imbens, Michael I. Jordan
arXiv:2606. 26418v1 Announce Type: new Abstract: A non-agentic "oracle" AI that estimates probabilities of future events faces a self-reference problem: once its answer is learned and acted upon, it can change the very probability it was asked to report.
By Jobst Heitzig
arXiv:2607. 14545v1 Announce Type: new Abstract: Machine-learned predictions can speed up offline NP-hard optimization, but asking a predictor what to do amounts to asking it to solve the problem, and committing an unchecked prediction forfeits every worst-case guarantee.
By Haifeng Li, Mo Hai