mlsanityData to 14 Oct 2026

Neural Interactive Proofs

arXiv 2412.08897 ↗First posted 12 Dec 2024cs.AI cs.LGcited by 7

Not on the map. Why?

Abstract (arXiv)

We consider the problem of how a trusted, but computationally bounded agent (a 'verifier') can learn to interact with one or more powerful but untrusted agents ('provers') in order to solve a given task. More specifically, we study the case in which agents are represented using neural networks and refer to solutions of this problem as neural interactive proofs. First we introduce a unifying framework based on prover-verifier games, which generalises previously proposed interaction protocols. We then describe several new protocols for generating neural interactive proofs, and provide a theoretical comparison of both new and existing approaches. Finally, we support this theory with experiments in two domains: a toy graph isomorphism problem that illustrates the key ideas, and a code validation task using large language models. In so doing, we aim to create a foundation for future work on neural interactive proofs and their application in building safer AI systems.

Key facts. Cited by 7 papers in the map (data through 14 Oct 2026). Builds on 16 of its 37 references that are in the map. First posted on arXiv 12 Dec 2024.

Builds on

16 of 37 references.

References · 16

  1. 2016Concrete Problems in AI Safetycited by 1,642
  2. 2023Weak-to-Strong Generalization: Eliciting Strong Capabilities With Weak Supervisioncited by 467
  3. 2018AI safety via debatecited by 400
  4. 2018Knockoff Nets: Stealing Functionality of Black-Box Modelscited by 292
  5. 2022Measuring Progress on Scalable Oversight for Large Language Modelscited by 225
  6. 2017Learning with Opponent-Learning Awarenesscited by 222
  7. 2023An Overview of Catastrophic AI Riskscited by 222
  8. 2021Proof-of-Learning: Definitions and Practicecited by 84
  9. 2021Introduction to Neural Network Verificationcited by 63
  10. 2023Scalable AI Safety via Doubly-Efficient Debatecited by 57
  11. 2017SafetyNets: Verifiable Execution of Deep Neural Networks on an Untrusted Cloudcited by 50
  12. 2023Managing extreme AI risks amid rapid progresscited by 48
4 more
  1. 2019Differentiable Game Mechanicscited by 43
  2. 2022Machine Learning Model Sizes and the Parameter Gapcited by 39
  3. 2021Learning to Give Checkable Answers with Prover-Verifier Gamescited by 32
  4. 2021Modeling Precomputation In Games Played Under Computational Constraintscited by 1

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2025: 3; 2026: 4.

7 papers.

  1. 2025Open Challenges in Multi-Agent Security: Towards Secure Systems of Interacting AI Agentscited by 71
  2. 2025Avoiding Obfuscation with Prover-Estimator Debatecited by 14
  3. 2026Trust but Verify: Prover-Verifier Deliberation for Selective LLM Predictioncited by 4
  4. 2026How to Avoid Debate: Scalable AI Safety via Doubly-Efficient Interactive Proofscited by 2
  5. 2025Bayesian Decision Making around Expertscited by 0
  6. 2026Human-LLM Deliberation as Interactive Proof: Conditions for Verifiability Without Transparencycited by 0
  7. 2026How to Have a Sensitive Debate: An Instance-Optimal Protocol for AI Debatecited by 0