[KongchangAI]
BenchmarkTHEATRON Benchmark

THEATRON

A benchmark for evaluating the persistence of personality in LLMs during long-horizon decision making. It uses financial replay as a controlled experimental environment and, through behavioral-finance personality injection and multiple metrics (intra-personality stability, cross-personality separability, and action collapse), quantifies whether personality continues to influence behavior across a sequence of decisions.

Timeline (last 90 days)

Nov 10

THEATRON is a benchmark for evaluating persona persistence in LLM long-horizon decision-making.

Unverified50%
Nov 10

THEATRON uses financial replay as a controlled experimental environment, in which models are assigned fixed behavioral-finance personas, receive account-state feedback relevant to their own trajectories under shared daily market evidence, and make continuous decisions over trajectories spanning several months.

Unverified50%
Nov 10

The evaluation metrics focused on cumulative returns and action trajectories, and on this basis diagnosed three dimensions: intra-persona stability, cross-persona separability and action collapse.

Unverified50%
Nov 10

These behavioral shifts cannot reliably form coherent persona-specific strategies.

Unverified50%
Nov 10

Apparently stable discriminability may stem from stereotyped action collapse, while apparently diverse strategies may be accompanied by noisy trajectory switching.

Unverified50%
Nov 10

The research team released the code and benchmark data on GitHub (yuyan12138/theatron).

Unverified50%

All Facts (6)

Source Articles