Image de couverture de livre

Proof-of-Use: Mitigating Tool-Call Hacking in Deep Research Agents

Fuente: arXiv

Enregistré dans:

Détails bibliographiques
Auteurs principaux:	Ma, SHengjie, Deng, Chenlong, Mao, Jiaxin, Huang, Jiadeng, Wang, Teng, Wu, Junjie, Zhang, Changwang, wang, Jun
Format:	Preprint
Publié:	2025
Sujets:	Artificial Intelligence
Accès en ligne:	Acceder al recurso
Tags:	Ajouter un tag Pas de tags, Soyez le premier à ajouter un tag!

Documents similaires

W&D:Scaling Parallel Tool Calling for Efficient Deep Research Agents
par: Lin, Xiaoqiang, et autres
Publié: (2026)

UniToolCall: Unifying Tool-Use Representation, Data, and Evaluation for LLM Agents
par: Liang, Yijuan, et autres
Publié: (2026)

LawThinker: A Deep Research Legal Agent in Dynamic Environments
par: Yang, Xinyu, et autres
Publié: (2026)

Reward Hacking Benchmark: Measuring Exploits in LLM Agents with Tool Use
par: Thaman, Kunvar
Publié: (2026)

Reinforced Agent: Inference-Time Feedback for Tool-Calling Agents
par: Ta, Anh, et autres
Publié: (2026)

DeepImageSearch: Benchmarking Multimodal Agents for Context-Aware Image Retrieval in Visual Histories
par: Deng, Chenlong, et autres
Publié: (2026)

SAGE: Strategy-Adaptive Generation Engine for Query Rewriting
par: Wang, Teng, et autres
Publié: (2025)

IR$^3$: Contrastive Inverse Reinforcement Learning for Interpretable Detection and Mitigation of Reward Hacking
par: Beigi, Mohammad, et autres
Publié: (2026)

OThink-R1: Intrinsic Fast/Slow Thinking Mode Switching for Over-Reasoning Mitigation
par: Zhang, Shengjia, et autres
Publié: (2025)

PhotoBench: Beyond Visual Matching Towards Personalized Intent-Driven Photo Retrieval
par: Xu, Tianyi, et autres
Publié: (2026)

OThink-SRR1: Search, Refine and Reasoning with Reinforced Learning for Large Language Models
par: Liang, Haijian, et autres
Publié: (2026)

AFSPP: Agent Framework for Shaping Preference and Personality with Large Language Models
par: He, Zihong, et autres
Publié: (2024)

To Call or Not to Call: A Framework to Assess and Optimize LLM Tool Calling
par: Wu, Qinyuan, et autres
Publié: (2026)

Adversarial Reward Auditing for Active Detection and Mitigation of Reward Hacking
par: Beigi, Mohammad, et autres
Publié: (2026)

ODIN: Disentangled Reward Mitigates Hacking in RLHF
par: Chen, Lichang, et autres
Publié: (2024)

Assertion-Conditioned Compliance: A Provenance-Aware Vulnerability in Multi-Turn Tool-Calling Agents
par: Waqas, Daud, et autres
Publié: (2025)

Mitigating Preference Hacking in Policy Optimization with Pessimism
par: Gupta, Dhawal, et autres
Publié: (2025)

OrchJail: Jailbreaking Tool-Calling Text-to-Image Agents by Orchestration-Guided Fuzzing
par: Chen, Jianming, et autres
Publié: (2026)

Interleaved Tool-Call Reasoning for Protein Function Understanding
par: Fan, Chuanliu, et autres
Publié: (2026)

ECG-Agent: On-Device Tool-Calling Agent for ECG Multi-Turn Dialogue
par: Chung, Hyunseung, et autres
Publié: (2026)

Hacking CTFs with Plain Agents
par: Turtayev, Rustem, et autres
Publié: (2024)

From Allies to Adversaries: Manipulating LLM Tool-Calling through Adversarial Injection
par: Wang, Haowei, et autres
Publié: (2024)

Reward Hacking Mitigation using Verifiable Composite Rewards
par: Tarek, Mirza Farhan Bin, et autres
Publié: (2025)

Repairing Reward Functions with Feedback to Mitigate Reward Hacking
par: Hatgis-Kessell, Stephane, et autres
Publié: (2025)

AWARE-US: Preference-Aware Infeasibility Resolution in Tool-Calling Agents
par: Kurmaz, Mehmet
Publié: (2026)

EigentSearch-Q+: Enhancing Deep Research Agents with Structured Reasoning Tools
par: Zhang, Boer, et autres
Publié: (2026)

OSCAR: Optimization-Steered Agentic Planning for Composed Image Retrieval
par: Wang, Teng, et autres
Publié: (2026)

FinToolBench: Evaluating LLM Agents for Real-World Financial Tool Use
par: Lu, Jiaxuan, et autres
Publié: (2026)

Reward Shaping to Mitigate Reward Hacking in RLHF
par: Fu, Jiayi, et autres
Publié: (2025)

SciToolAgent: A Knowledge Graph-Driven Scientific Agent for Multi-Tool Integration
par: Ding, Keyan, et autres
Publié: (2025)

MiniScope: A Least Privilege Framework for Authorizing Tool Calling Agents
par: Zhu, Jinhao, et autres
Publié: (2025)

Multi-Turn Reinforcement Learning for Tool-Calling Agents with Iterative Reward Calibration
par: Modecrua, Wachiravit, et autres
Publié: (2026)

ASA: Training-Free Representation Engineering for Tool-Calling Agents
par: Wang, Youjin, et autres
Publié: (2026)

Ghost Tool Calls: Issue-Time Privacy for Speculative Agent Tools
par: Mohammadi, Bardia, et autres
Publié: (2026)

ToolCUA: Towards Optimal GUI-Tool Path Orchestration for Computer Use Agents
par: Hu, Xuhao, et autres
Publié: (2026)

RewardHackingAgents: Benchmarking Evaluation Integrity for LLM ML-Engineering Agents
par: Atinafu, Yonas, et autres
Publié: (2026)

LLM Agents can Autonomously Hack Websites
par: Fang, Richard, et autres
Publié: (2024)

DeepResearch-9K: A Challenging Benchmark Dataset of Deep-Research Agent
par: Wu, Tongzhou, et autres
Publié: (2026)

Budget-Aware Tool-Use Enables Effective Agent Scaling
par: Liu, Tengxiao, et autres
Publié: (2025)

LLM Agent Honeypot: Monitoring AI Hacking Agents in the Wild
par: Reworr, et autres
Publié: (2024)