paper-with-me

홈 › Papers

Gaming Tool Preferences in Agentic LLMs

2025-05-23 · Kazem Faghih, Wenxiao Wang, Yize Cheng, Siddhant Bharti, Gaurang Sriramanan, Sriram Balasubramanian, Parsa Hosseini, Soheil Feizi

Large language models (LLMs) can now access a wide range of external tools, thanks to the Model Context Protocol (MCP). This greatly expands their abilities as various agents. However, LLMs rely entirely on the text descriptions of tools to decide which ones to use--a process that is surprisingly fragile. In this work, we expose a vulnerability in prevalent tool/function-calling protocols by investigating a series of edits to tool descriptions, some of which can drastically increase a tool's usage from LLMs when competing with alternatives. Through controlled experiments, we show that tools with properly edited descriptions receive over 10 times more usage from GPT-4.1 and Qwen2.5-7B than tools with original descriptions. We further evaluate how various edits to tool descriptions perform when competing directly with one another and how these trends generalize or differ across a broader set of 10 different models. These phenomenons, while giving developers a powerful way to promote their tools, underscore the need for a more reliable foundation for agentic LLMs to select and utilize tools and resources.

📄 PDF Abstract BibTeX arXiv:2505.18135

Code (1)

kazemf78/Gaming-Tool-Preferences 공식 구현

Methods 이 논문이 사용한 방법론

Attention 설명 없음
Linear Layer A Linear Layer is a projection $\mathbf{XW + b}$.
Dense Connections Dense Connections, or Fully Connected Connections, are a type of layer in a deep neural network that use a linear operation where every input is connected to every output…
Softmax The Softmax output function transforms a previous layer's output into a vector of probabilities. It is commonly used for multiclass classification. Given an input vector $x$…
Position-Wise Feed-Forward Layer 설명 없음
Absolute Position Encodings Absolute Position Encodings are a type of position embeddings for [Transformer-based models] where positional encodings are…
Label Smoothing Label Smoothing is a regularization technique that introduces noise for the labels. This accounts for the fact that datasets may have mistakes in them, so maximizing the…
Multi-Head Attention 설명 없음

Similar Papers 제목 키워드 기반

HotelQuEST: Balancing Quality and Efficiency in Agentic Search

2026-02-27 · Guy Hadad, Shadi Iskander, Oren Kalinsky, Sofia Tolmach 외 arxiv

Agentic search has emerged as a promising paradigm for adaptive retrieval systems powered by large language models (LLMs). However, existing benchmarks primarily focus on quality, overlooking efficiency factors that are …

INTENT-AS-A-TOOL Makes it Easy to Track Agentic Misalignment

2026-08-27 · Yutong Zhang, Jianshuo Dong, Peng Xu, Long Wang 외 arxiv

As large language models (LLMs) are deployed as autonomous agents, safety failures increasingly involve consequential actions. We study agentic misalignment, where agents take harmful actions under goal conflicts and pre…

ToolRM: Towards Agentic Tool-Use Reward Modeling

2025-10-30 · Renhao Li, Jianhong Tu, Yang Su, Yantao Liu 외 arxiv

Reward models (RMs) play a critical role in aligning large language models (LLMs) with human preferences. Yet in the domain of tool learning, the lack of RMs specifically designed for function-calling tasks has limited p…

Sentiment Analysis On YouTube Comments Using Machine Learning Techniques Based On Video Games Content

2025-11-10 · Adi Danish Bin Muhammad Amin, Mohaiminul Islam Bhuiyan, Nur Shazwani Kamarudin, Zulfahmi Toh 외 arxiv

The rapid evolution of the gaming industry, driven by technological advancements and a burgeoning community, necessitates a deeper understanding of user sentiments, especially as expressed on popular social media platfor…

Sentiment Analysis

AgenticRec: A Recommendation-Oriented Agentic Framework with Progressive Tool-Integrated Reasoning Optimization

2026-03-23 · Tianyi Li, Zixuan Wang, Guidong Lei, Xiaodong Li 외 arxiv

Recommender agents built on Large Language Models offer a promising paradigm for personalized recommendation. However, existing agents typically suffer from a misalignment between their tool-integrated reasoning trajecto…