Saved in:
Bibliographic Details
Main Authors: Zhang, Erica, Zhang, Fangzhao, Pappu, Aneesh, El, Batu, Blanchet, Jose, Athey, Susan, Liu, Jiashuo, Zou, James
Format: Preprint
Published: 2026
Subjects:
Online Access:https://arxiv.org/abs/2605.13909
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866911683378675712
author Zhang, Erica
Zhang, Fangzhao
Pappu, Aneesh
El, Batu
Blanchet, Jose
Athey, Susan
Liu, Jiashuo
Zou, James
author_facet Zhang, Erica
Zhang, Fangzhao
Pappu, Aneesh
El, Batu
Blanchet, Jose
Athey, Susan
Liu, Jiashuo
Zou, James
contents Negotiation is a central mechanism of economic exchange, shaping markets, procurement, labor agreements, and resource allocation. It is also a canonical testbed for agentic language models, requiring multi-turn interaction under hidden preferences, strategic communication, and binding constraints. These properties make negotiation hard to evaluate: unlike math or code, it has no intrinsic verifier. Existing LLM negotiation evaluations rely on LLM-vs.-LLM interaction or aggregate outcomes such as deal rate, leaving failures opaque. We introduce Terms-Bench, short for Testbed for Economic Reasoning in Multi-turn Strategy, a Bayesian-game framework that makes the environment itself the verifier by specifying the counterpart's latent type, policy, and payoff structure. We instantiate it in bilateral price negotiation, where the counterpart's private state and simulator policy are hidden from the agent but observable to the evaluator. This turns the counterpart from a black-box opponent into a diagnostic instrument, enabling agent-attributable failure analysis and oracle-reference optimality gaps. Evaluating 13 LLM agents spanning frontier systems from major providers, Terms-Bench turns negotiation evaluation from aggregate ranking into actionable diagnosis: where agents fail, why they fail, and what to strengthen. Empirically, frontier models saturate deal rate yet diverge in surplus extraction, cue use, belief calibration, and compliance, revealing agent-specific bargaining bottlenecks masked by prior benchmarks.
format Preprint
id arxiv_https___arxiv_org_abs_2605_13909
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle TERMS-Bench: Diagnosing LLM Negotiation Agents Beyond Deal Rate
Zhang, Erica
Zhang, Fangzhao
Pappu, Aneesh
El, Batu
Blanchet, Jose
Athey, Susan
Liu, Jiashuo
Zou, James
Computer Science and Game Theory
Artificial Intelligence
Negotiation is a central mechanism of economic exchange, shaping markets, procurement, labor agreements, and resource allocation. It is also a canonical testbed for agentic language models, requiring multi-turn interaction under hidden preferences, strategic communication, and binding constraints. These properties make negotiation hard to evaluate: unlike math or code, it has no intrinsic verifier. Existing LLM negotiation evaluations rely on LLM-vs.-LLM interaction or aggregate outcomes such as deal rate, leaving failures opaque. We introduce Terms-Bench, short for Testbed for Economic Reasoning in Multi-turn Strategy, a Bayesian-game framework that makes the environment itself the verifier by specifying the counterpart's latent type, policy, and payoff structure. We instantiate it in bilateral price negotiation, where the counterpart's private state and simulator policy are hidden from the agent but observable to the evaluator. This turns the counterpart from a black-box opponent into a diagnostic instrument, enabling agent-attributable failure analysis and oracle-reference optimality gaps. Evaluating 13 LLM agents spanning frontier systems from major providers, Terms-Bench turns negotiation evaluation from aggregate ranking into actionable diagnosis: where agents fail, why they fail, and what to strengthen. Empirically, frontier models saturate deal rate yet diverge in surplus extraction, cue use, belief calibration, and compliance, revealing agent-specific bargaining bottlenecks masked by prior benchmarks.
title TERMS-Bench: Diagnosing LLM Negotiation Agents Beyond Deal Rate
topic Computer Science and Game Theory
Artificial Intelligence
url https://arxiv.org/abs/2605.13909