Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Pollo, Jorge Carrasco, Kapetangeorgis, Ioannis, Rosenthal, Joshua, Yao, John Hua
Format: Preprint
Veröffentlicht: 2026
Schlagworte:
Online-Zugang:https://arxiv.org/abs/2602.18230
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866910029848772608
author Pollo, Jorge Carrasco
Kapetangeorgis, Ioannis
Rosenthal, Joshua
Yao, John Hua
author_facet Pollo, Jorge Carrasco
Kapetangeorgis, Ioannis
Rosenthal, Joshua
Yao, John Hua
contents Large Language Models (LLMs) demonstrate significant potential in multi-agent negotiation tasks, yet evaluation in this domain remains challenging due to a lack of robust and generalizable benchmarks. Abdelnabi et al. (2024) introduce a negotiation benchmark based on Scoreable Games, with the aim of developing a highly complex and realistic evaluation framework for LLMs. Our work investigates the reproducibility of claims in their benchmark, and provides a deeper understanding of its usability and generalizability. We replicate the original experiments on additional models, and introduce additional metrics to verify negotiation quality and evenness of evaluation. Our findings reveal that while the benchmark is indeed complex, model comparison is ambiguous, raising questions about its objectivity. Furthermore, we identify limitations in the experimental setup, particularly in information leakage detection and thoroughness of the ablation study. By examining and analyzing the behavior of a wider range of models on an extended version of the benchmark, we reveal insights that provide additional context to potential users. Our results highlight the importance of context in model-comparative evaluations.
format Preprint
id arxiv_https___arxiv_org_abs_2602_18230
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle [Re] Benchmarking LLM Capabilities in Negotiation through Scoreable Games
Pollo, Jorge Carrasco
Kapetangeorgis, Ioannis
Rosenthal, Joshua
Yao, John Hua
Machine Learning
Artificial Intelligence
Large Language Models (LLMs) demonstrate significant potential in multi-agent negotiation tasks, yet evaluation in this domain remains challenging due to a lack of robust and generalizable benchmarks. Abdelnabi et al. (2024) introduce a negotiation benchmark based on Scoreable Games, with the aim of developing a highly complex and realistic evaluation framework for LLMs. Our work investigates the reproducibility of claims in their benchmark, and provides a deeper understanding of its usability and generalizability. We replicate the original experiments on additional models, and introduce additional metrics to verify negotiation quality and evenness of evaluation. Our findings reveal that while the benchmark is indeed complex, model comparison is ambiguous, raising questions about its objectivity. Furthermore, we identify limitations in the experimental setup, particularly in information leakage detection and thoroughness of the ablation study. By examining and analyzing the behavior of a wider range of models on an extended version of the benchmark, we reveal insights that provide additional context to potential users. Our results highlight the importance of context in model-comparative evaluations.
title [Re] Benchmarking LLM Capabilities in Negotiation through Scoreable Games
topic Machine Learning
Artificial Intelligence
url https://arxiv.org/abs/2602.18230