Cover Image

Tool-Call Dependency Structure is Linearly Decodable in LLM Agent Residual Streams

Fuente: arXiv

Saved in:

Bibliographic Details
Main Authors:	Sun, Tianda, Kazakov, Dimitar
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

Peak-Then-Collapse and the Four Interface Channels of Knowledge-Graph Tool Use
by: Sun, Tianda, et al.
Published: (2026)

Kinship Data Benchmark for Multi-hop Reasoning
by: Sun, Tianda, et al.
Published: (2026)

Mitigating Bias in Text Classification via Prompt-Based Text Transformation
by: Barker, Charmaine, et al.
Published: (2023)

SimpleTool: Parallel Decoding for Real-Time LLM Function Calling
by: Shi, Xiaoxin, et al.
Published: (2026)

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

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

ToolSpec: Accelerating Tool Calling via Schema-Aware and Retrieval-Augmented Speculative Decoding
by: Xia, Heming, et al.
Published: (2026)

MeNTi: Bridging Medical Calculator and LLM Agent with Nested Tool Calling
by: Zhu, Yakun, et al.
Published: (2024)

MemTool: Optimizing Short-Term Memory Management for Dynamic Tool Calling in LLM Agent Multi-Turn Conversations
by: Lumer, Elias, et al.
Published: (2025)

From Text to Voice: A Reproducible and Verifiable Framework for Evaluating Tool Calling LLM Agents
by: Laskar, Md Tahmid Rahman, et al.
Published: (2026)

Decodable but Not Corrected by Fixed Residual-Stream Linear Steering: Evidence from Medical LLM Failure Regimes
by: Liu, Ming
Published: (2026)

ToolFlow: Boosting LLM Tool-Calling Through Natural and Coherent Dialogue Synthesis
by: Wang, Zezhong, et al.
Published: (2024)

When2Call: When (not) to Call Tools
by: Ross, Hayley, et al.
Published: (2025)

Natural Language Tools: A Natural Language Approach to Tool Calling In Large Language Agents
by: Johnson, Reid T., et al.
Published: (2025)

TOOL-ED: Enhancing Empathetic Response Generation with the Tool Calling Capability of LLM
by: Cao, Huiying, et al.
Published: (2024)

ContextWeaver: Selective and Dependency-Structured Memory Construction for LLM Agents
by: Wu, Yating, et al.
Published: (2026)

ToolWeave: Structured Synthesis of Complex Multi-Turn Tool-Calling Dialogues
by: Khandelwal, Dinesh, et al.
Published: (2026)

AnyTool: Self-Reflective, Hierarchical Agents for Large-Scale API Calls
by: Du, Yu, et al.
Published: (2024)

Ghost Tool Calls: Issue-Time Privacy for Speculative Agent Tools
by: Mohammadi, Bardia, et al.
Published: (2026)

Tool Calling is Linearly Readable and Steerable in Language Models
by: Wu, Zekun, et al.
Published: (2026)

Tool-to-Agent Retrieval: Bridging Tools and Agents for Scalable LLM Multi-Agent Systems
by: Lumer, Elias, et al.
Published: (2025)

ToolACE: Winning the Points of LLM Function Calling
by: Liu, Weiwen, et al.
Published: (2024)

SCRIBE: Structured Chain Reasoning for Interactive Behaviour Explanations using Tool Calling
by: Fawzi, Fares, et al.
Published: (2025)

LoopTool: Closing the Data-Training Loop for Robust LLM Tool Calls
by: Zhang, Kangning, et al.
Published: (2025)

Efficient Tool-Calling Multi-Expert NPC Agent for Commonsense Persona-Grounded Dialogue
by: Nuriyev, Mahammad
Published: (2025)

Residual Matrix Transformers: Scaling the Size of the Residual Stream
by: Mak, Brian, et al.
Published: (2025)

Do Agents Need to Plan Step-by-Step? Rethinking Planning Horizon in Data-Centric Tool Calling
by: Otani, Naoki, et al.
Published: (2026)

ToolRM: Outcome Reward Models for Tool-Calling Large Language Models
by: Agarwal, Mayank, et al.
Published: (2025)

Break the Sequential Dependency of LLM Inference Using Lookahead Decoding
by: Fu, Yichao, et al.
Published: (2024)

The Landscape of Emerging AI Agent Architectures for Reasoning, Planning, and Tool Calling: A Survey
by: Masterman, Tula, et al.
Published: (2024)

Building a Macedonian Recipe Dataset: Collection, Parsing, and Comparative Analysis
by: Sasanski, Darko, et al.
Published: (2025)

CoLT: Reasoning with Chain of Latent Tool Calls
by: Zhu, Fangwei, et al.
Published: (2026)

Alignment for Efficient Tool Calling of Large Language Models
by: Xu, Hongshen, et al.
Published: (2025)

Decoder-only Streaming Transformer for Simultaneous Translation
by: Guo, Shoutao, et al.
Published: (2024)

Assertion-Conditioned Compliance: A Provenance-Aware Vulnerability in Multi-Turn Tool-Calling Agents
by: Waqas, Daud, et al.
Published: (2025)

Uncertainty Quantification for LLM Function-Calling
by: Ye, Zihuiwen, et al.
Published: (2026)

An LLM Compiler for Parallel Function Calling
by: Kim, Sehoon, et al.
Published: (2023)

Analysing the Residual Stream of Language Models Under Knowledge Conflicts
by: Zhao, Yu, et al.
Published: (2024)

Not All Proofs Are Equal: Evaluating LLM Proof Quality Beyond Correctness
by: Petrov, Ivo, et al.
Published: (2026)

Trajectory2Task: Training Robust Tool-Calling Agents with Synthesized Yet Verifiable Data for Complex User Intents
by: Wang, Ziyi, et al.
Published: (2026)