[KongchangAI]
Concept评估侧部分可观测扩展式博弈

e-PoG

e-PoG is a framework that formalizes AI agent deception mitigation as an evaluation-side partially observable extensive-form game. Through a max-min game structure, the monitor is made to expose deceptive behavior, and the subgame perfect oversight equilibrium (SPOE) is defined as the solution concept.

Timeline (last 90 days)

Nov 10

The researchers formalize AI agent deception mitigation as an evaluation-side partially observable extensive-form game (e-PoG), a max-min framework in which the agent maximizes reward and the monitor exposes deception from complete interaction trajectories to minimize agent payoff, with a fixed verifier validating the monitor's claims and converting them into penalties.

Unverified50%
Nov 10

The e-PoG framework defines a solution concept—subgame perfect oversight equilibrium (SPOE)—and the paper proves the existence of SPOE in finite games.

Unverified50%
Nov 10

Experimental results show that e-PoG outperforms the strongest baseline by 26.0% on the deception mitigation metric and improves task success rate by up to 39.8%.

Unverified50%
Nov 10

The gains of e-PoG persist under weak-to-strong supervision.

Unverified50%
Nov 10

The code for e-PoG is open-sourced at https://github.com/Toyoid/deceptive-agent.

Unverified50%
Nov 10

The effectiveness of e-PoG depends on verifier quality, and building a reliable verifier is itself a challenge in practical applications.

Unverified50%
Nov 10

The experimental scale of e-PoG is limited, and its generalization ability needs to be tested in more complex open scenarios in the future.

Unverified50%
Nov 10

In e-PoG, the monitor's claims are checked for consistency against evidence in the interaction trajectory via NLI verification, while structured facts such as tool-call parameters and environment state changes are verified via determinism checks.

Unverified50%

All Facts (8)

Source Articles