pith. sign in

logπθ y+t π′t y+t −logπθ y−t π′t y−t −η 2 #2 , whereπ′t= argminπ E(x,y+t ,y−t)∼Dt

1 Pith paper cite this work. Polarity classification is still indexing.

1 Pith paper citing it

fields

cs.AI 1

years

2025 1

verdicts

UNVERDICTED 1

representative citing papers

Multiplayer Nash Preference Optimization

cs.AI · 2025-09-27 · unverdicted · novelty 6.0

MNPO extends NLHF to multiplayer Nash games, inheriting equilibrium guarantees while showing empirical gains on instruction-following benchmarks under diverse preferences.

citing papers explorer

Showing 1 of 1 citing paper.

  • Multiplayer Nash Preference Optimization cs.AI · 2025-09-27 · unverdicted · none · ref 47

    MNPO extends NLHF to multiplayer Nash games, inheriting equilibrium guarantees while showing empirical gains on instruction-following benchmarks under diverse preferences.