AI Paper Atlas

Research map / Large language models / Alignment and preferences

Multiplayer Nash Preference Optimization

Fang Wu, Xu Huang, Weihao Xuan, Zhiwei Zhang, Yijia Xiao, Guancheng Wan, Xiaomin Li, Bing Hu, Peng Xia, Jure Leskovec, Yejin Choi

ICLR 2026 · Large language models › Alignment and preferences · approach: Reward modeling for diverse preferences

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
increase number of players in TD-MNPO from 1 to 3 (ablation)TD-MNPO · AlpacaEval 2.0 (LC WR)LC win rate: 53.32 -> 57.27Improved
increase number of players from 3 to 4 (ablation)TD-MNPO · AlpacaEval 2.0 (LC WR)LC win rate: 57.27 -> 57.42No clear effect
use 3 players instead of 2 in HT-MNPO (Athene-RM-8B) (ablation)HT-MNPO · AlpacaEval 2.0 (LC WR)LC win rate: 55.71 -> 59.64Improved
use 3 players instead of 2 in HT-MNPO (ArmoRM-Llama3) (ablation)HT-MNPO · AlpacaEval 2.0 (LC WR)LC win rate: 55.43 -> 57.63Improved
use 3 players instead of 2 in HT-MNPO (Skywork-Reward-V2) (ablation)HT-MNPO · AlpacaEval 2.0 (LC WR)LC win rate: 54.25 -> 56.01Improved
replace INPO with TD-MNPO on Llama-3-8B-it (ablation)Llama-3-8B-it · AlpacaEval 2.0 (LC WR)LC win rate: 41.48 -> 42.94Improved
replace INPO with TD-MNPO on Gemma-2-9B-itGemma-2-9B-it · Arena-Hard (WR)win rate: 48.03 -> 52.26Improved
replace INPO with TD-MNPO on Gemma-2-9B-itGemma-2-9B-it · AlpacaEval 2.0 (LC WR)LC win rate: 56.09 -> 57.27Improved
use TD-MNPO instead of INPO under different judges (GPT-4-1106-preview)Gemma-2-9B-it · AlpacaEval 2.0 (3 runs)LC win rate: 49.30 -> 54.05Improved
use TD-MNPO instead of baselines on academic benchmarksGemma-2-9B-it · 7 instruction/knowledge/commonsense benchmarksaverage accuracy: 70.28 -> 71.08Improved
use TD-MNPO instead of SimPO regarding TruthfulQA degradationGemma-2-9B-it · TruthfulQAaccuracy: 63.40 -> 70.26Improved
use TD-MNPO instead of baselines on math/codingGemma-2-9B-it · Math and coding averageaverage score: 47.33 -> 48.10Improved

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy