[KongchangAI]
Conceptnative decision models

Native Decision Model

A model designed specifically for sequential decision-making or planning tasks, typically based on reinforcement learning or search. It is trained directly in the state-action space and outputs executable plans or action sequences.

Timeline (last 90 days)

Nov 5

Native decision models refer to models designed specifically for sequential decision-making or planning tasks (such as planners based on reinforcement learning or search), as distinct from general-purpose language models (such as the GPT series).

Unverified50%
Nov 5

Native decision models achieve 62.50% accuracy on the TandemBench benchmark, but are close to random on optimality verification.

Unverified50%
Nov 5

The study evaluated 40 native decision models and compared them with 10 general-purpose language models.

Unverified50%
Nov 5

In the optimality verification task, native decision models perform close to the random level.

Unverified50%

All Facts (4)

Source Articles