Model techniques map
Techniquespost-trainingreward modelling

specific method · filed under post-training

Abstention-aware reward for factual QA

Rewards answering factual questions when confidence is sufficient while allowing abstention or hedged guesses when it is not.

Also called Abstention-aware rewards for factual QA.

source
1
model
1
lab adopt it
1
strongest
used

How sources treat it

One count per evidence span, weakest treatment to strongest.

used 1

Documented in

Evidence

1 span quoted from the sources, strongest treatment first.

The largest is short-form factual QA with abstention-aware rewards: answering only pays off when the model is likely to be right, so the optimal policy is to answer when confident and otherwise say “I don’t know” or give a hedged best guess.

usedpost trainingin InklingThinking Machines Lab

Filed alongside

Other methods under post-training :: reward modelling.