Cover Image

When Agents Look the Same: Quantifying Distillation-Induced Similarity in Tool-Use Behaviors

Fuente: arXiv

Saved in:

Bibliographic Details
Main Authors:	Yang, Chenghao, Zhang, Yuning, Wen, Zhoufutu, Gong, Tao, Liu, Jiaheng, Chu, Qi, Yu, Nenghai
Format:	Preprint
Published:	2026
Subjects:	Computation and Language
Online Access:	Acceder al recurso
Tags:	Add Tag No Tags, Be the first to tag this record!

Similar Items

MARS-Bench: A Multi-turn Athletic Real-world Scenario Benchmark for Dialogue Evaluation
by: Yang, Chenghao, et al.
Published: (2025)

Quantification of Large Language Model Distillation
by: Lee, Sunbowen, et al.
Published: (2025)

Llama SLayer 8B: Shallow Layers Hold the Key to Knowledge Injection
by: Chen, Tianxiang, et al.
Published: (2024)

Flora: Effortless Context Construction to Arbitrary Length and Scale
by: Chen, Tianxiang, et al.
Published: (2025)

The Evolution of Tool Use in LLM Agents: From Single-Tool Call to Multi-Tool Orchestration
by: Xu, Haoyuan, et al.
Published: (2026)

The Same But Different: Structural Similarities and Differences in Multilingual Language Modeling
by: Zhang, Ruochen, et al.
Published: (2024)

Tools in the Loop: Quantifying Uncertainty of LLM Question Answering Systems That Use Tools
by: Lymperopoulos, Panagiotis, et al.
Published: (2025)

The Confidence Dichotomy: Analyzing and Mitigating Miscalibration in Tool-Use Agents
by: Xuan, Weihao, et al.
Published: (2026)

TopoCurate:Modeling Interaction Topology for Tool-Use Agent Training
by: Yang, Jinluan, et al.
Published: (2026)

The Tool Illusion: Rethinking Tool Use in Web Agents
by: Lou, Renze, et al.
Published: (2026)

Same Company, Same Signal: The Role of Identity in Earnings Call Transcripts
by: Yu, Ding, et al.
Published: (2024)

OmniRAG-Agent: Agentic Omnimodal Reasoning for Low-Resource Long Audio-Video Question Answering
by: Zhu, Yifan, et al.
Published: (2026)

When Hindsight is Not 20/20: Testing Limits on Reflective Thinking in Large Language Models
by: Li, Yanhong, et al.
Published: (2024)

When Meaning Stays the Same, but Models Drift: Evaluating Quality of Service under Token-Level Behavioral Instability in LLMs
by: Li, Xiao, et al.
Published: (2025)

GTA-2: Benchmarking General Tool Agents from Atomic Tool-Use to Open-Ended Workflows
by: Wang, Jize, et al.
Published: (2026)

When Benign Inputs Lead to Severe Harms: Eliciting Unsafe Unintended Behaviors of Computer-Use Agents
by: Jones, Jaylen, et al.
Published: (2026)

TableMind++: An Uncertainty-Aware Programmatic Agent for Tool-Augmented Table Reasoning
by: Cheng, Mingyue, et al.
Published: (2026)

Re-Initialization Token Learning for Tool-Augmented Large Language Models
by: Li, Chenghao, et al.
Published: (2025)

FinToolSyn: A forward synthesis Framework for Financial Tool-Use Dialogue Data with Dynamic Tool Retrieval
by: Huang, Caishuang, et al.
Published: (2026)

Chain-of-Agents: End-to-End Agent Foundation Models via Multi-Agent Distillation and Agentic RL
by: Li, Weizhen, et al.
Published: (2025)

Context-Aware Weakly Supervised Image Manipulation Localization with SAM Refinement
by: Wang, Xinghao, et al.
Published: (2025)

Value-Spectrum: Quantifying Preferences of Vision-Language Models via Value Decomposition in Social Media Contexts
by: Li, Jingxuan, et al.
Published: (2024)

LLM Agents Already Know When to Call Tools -- Even Without Reasoning
by: Sun, Chung-En, et al.
Published: (2026)

Learning How to Use Tools, Not Just When: Pattern-Aware Tool-Integrated Reasoning
by: Xu, Ningning, et al.
Published: (2025)

Effective Distillation of Table-based Reasoning Ability from LLMs
by: Yang, Bohao, et al.
Published: (2023)

MMedAgent: Learning to Use Medical Tools with Multi-modal Agent
by: Li, Binxu, et al.
Published: (2024)

WorldTravel: A Realistic Multimodal Travel-Planning Benchmark with Tightly Coupled Constraints
by: Wang, Zexuan, et al.
Published: (2026)

ToolScope: Enhancing LLM Agent Tool Use through Tool Merging and Context-Aware Filtering
by: Liu, Marianne Menglin, et al.
Published: (2025)

Learning to Use Tools via Cooperative and Interactive Agents
by: Shi, Zhengliang, et al.
Published: (2024)

Look Before You Leap: Autonomous Exploration for LLM Agents
by: Ye, Ziang, et al.
Published: (2026)

QUDsim: Quantifying Discourse Similarities in LLM-Generated Text
by: Namuduri, Ramya, et al.
Published: (2025)

CCTU: A Benchmark for Tool Use under Complex Constraints
by: Ye, Junjie, et al.
Published: (2026)

When Thinking Backfires: Mechanistic Insights Into Reasoning-Induced Misalignment
by: Yan, Hanqi, et al.
Published: (2025)

Distilling LLM Agent into Small Models with Retrieval and Code Tools
by: Kang, Minki, et al.
Published: (2025)

Exploring Knowledge Purification in Multi-Teacher Knowledge Distillation for LLMs
by: Jin, Ruihan, et al.
Published: (2026)

MetaTool Benchmark for Large Language Models: Deciding Whether to Use Tools and Which to Use
by: Huang, Yue, et al.
Published: (2023)

CoTJudger: A Graph-Driven Framework for Automatic Evaluation of Chain-of-Thought Efficiency and Redundancy in LRMs
by: Li, Siyi, et al.
Published: (2026)

When to Trust Tools? Adaptive Tool Trust Calibration For Tool-Integrated Math Reasoning
by: Xu, Ruotao, et al.
Published: (2026)

RepMatch: Quantifying Cross-Instance Similarities in Representation Space
by: Modarres, Mohammad Reza, et al.
Published: (2024)

When the Same Coefficients Reach Different Places: Asymmetric Realizability in Transplanting Tokenizers across Large Language Models
by: Liu, Xiaoze, et al.
Published: (2025)