The paper investigates whether large language models (LLMs) correctly gauge their confidence when acting in a hidden‑information chess variant. In experiments where the location of a hidden royal piece is repeatedly relocated, the models’ stated probabilities about the piece’s position were almost never accurate at high confidence levels, with a calibration deficit concentrated in those high‑confidence events. Across multiple model configurations and providers, the same pattern emerged, and conventional evaluation metrics such as legality, cost, latency, and completion rate were found to be uncorrelated with belief quality, yet a model could still win the game despite poor confidence estimates.
By Bhushan Kashinath Joshi
The paper investigates why large language models (LLMs) struggle in strategic decision-making under incomplete information. It identifies two key gaps: an observation‑belief gap where LLMs’ internal representations of game states are accurate but brittle, and a belief‑action gap where converting these internal beliefs into actions is weak, leading to suboptimal payoffs. Experiments with Llama 3.1, Qwen3, and gpt‑oss confirm that acting optimally on decoded beliefs would improve outcomes in most games, highlighting a bottleneck in belief‑to‑action conversion.
By Jan Sobotka, Mustafa O. Karabag, Ufuk Topcu
The paper shows that large language model agents are far more likely to commit to a directional answer when presented with a professional-looking market panel, even if the panel’s data is fabricated. Across 12 frontier models, commitment rates jump from 6.5 % for a bare question to 54.0 % with evidence, and remain high (≈37 %) even when all numbers are invented. The study finds that the act/don’t‑act decision gate is the key failure point, can be trained to reduce false commitments, but is fragile to response format changes.
The study shows that large language model (LLM) agents are far more likely to commit to a directional prediction when presented with a professional‑looking market panel than when asked the same question directly, with commitment rates rising from 6.5% to 54.0% across 12 frontier models. Even when the panel’s data is entirely fabricated, commitment still increases significantly, indicating that the authority of the presentation, rather than the truth of the information, drives confident action. The authors demonstrate that this act/don’t‑act decision gate is narrow, model‑specific, and can be mitigated through supervised fine‑tuning, though its effectiveness depends on response format and context.
whyItMatters":"The findings reveal a specific vulnerability in LLMs where presentation style can override factual accuracy, highlighting the need for careful design and training to prevent misleading confidence in uncertain scenarios."
By Pranav Aggarwal
arXiv:2610.09371v1 Announce Type: cross
Abstract: Calibrated uncertainty quantification is essential to ensuring AI agents are trustworthy and reliable. However, when agents seek to maximize user eng...
By Raghu Arghal, Saswati Sarkar, Shirin Saeedi Bidokhti
arXiv:2607. 10814v1 Announce Type: cross Abstract: Evaluating LLM agents in hidden-information multi-agent settings is hard: final outcomes are high-variance and rarely reveal why an agent decided as it did.
By Yuan Gao, Jiangyi Yang, Yao Zhao, Yichi Zhang
arXiv:2609.22478v1 Announce Type: cross
Abstract: Behavioural evaluations of hosted language models can vary because the evaluated service, the measurement instrument, or both differ across runs. We...
By Bhushan Kashinath Joshi
The paper introduces XiangqiBench, an executable benchmark for evaluating large language model agents in Chinese chess. It tracks 8,568 multi‑turn trajectories from 12 frontier LLMs, revealing that metrics such as the Conversion Gap, Consistency Gap, and Simulation Gap overstate true closed‑loop success. The study shows that simply naming the correct move is insufficient; agents must reliably carry a plan through to a verified outcome against an opponent.
By Yekun Chai, Qiwei Peng, Haoyi Xiong
The paper investigates whether frozen language models can detect a corrupted reward signal by using a single verified record in a two‑option game. In the game, a payout swap and a lying reporter produce identical histories, but a single line confirming the true outcome allows the models to almost perfectly identify the liar. However, the models frequently misclassify honest reporters as liars, with error rates ranging from 26% to 58% depending on model size and wording, indicating a significant limitation in their ability to interpret verified data.
By Arman Nik Khah
The paper investigates whether confidence scores from a black-box decision model, Jev, truly reflect missing knowledge. Using over 15 public datasets and 6 synthetic task families, the authors find that while Jev’s confidence is calibrated on familiar closed-choice tasks, it fails to indicate when the model lacks relevant information—assigning high confidence to salient options even without answer-relevant data and overestimating accuracy on news beyond its knowledge boundary. Targeted yes/no questions about whether an outcome is settled or whether evidence suffices provide sharper indicators of knowledge gaps, but only when surface cues are controlled.
By Sharath M Shankaranarayana, Davor Runje, Jan Jannink
The paper argues that accuracy is an inadequate metric for certifying a coordinator’s theory‑of‑mind in multi‑agent systems, because the required accuracy depends on deployment‑specific costs and errors are often dependent. Experiments on task assignment, Level‑Based Foraging, Hanabi, and human judgments show that accuracy can both approve harmful language‑model deployments and reject beneficial ones. The authors propose a Blind‑Spot Gate that evaluates error costs, pairwise interactions, and dependence without deployment, achieving better safety outcomes than accuracy alone.
By Will Beaumaster, Paul Schrater
arXiv:2608. 04240v1 Announce Type: cross Abstract: Superhuman game engines in domains like chess have made expert-level evaluations easily accessible, yet they communicate what is true without the natural-language explanations that make such expertise educationally useful to experts and non-experts alike.
By S. Ashwin Hebbar, Peiyao Sheng, Sewoong Oh, Pramod Viswanath