LLM Agent Tool Generation

LLM: Large Language Model

Latest papers

11 papers1 paper in the last four weeks, against 2 the four weeks before. 0.0% of all new papers.

SchemaFill: Efficient LLM Tool Calling via Slot-Parallel Speculative Decoding
Oct 5, 2026Zhi-Kai Chen, Song-Yan Li, De-Chuan Zhan +1Speculative Decoding LLM Inference Acceleration
School of Artificial Intelligence, Nanjing University, China · National Key Laboratory for Novel Software Technology, Nanjing University, China · Nanjing University, China
VideoResearcher: Self-Improving Tool Design for Long-Video Understanding
Sep 17, 2026Dingqiang Ye, Dongdi Zhao, Kaishen Wang +11Tool-Using Vision-Language Agents Long-Video Understanding
Johns Hopkins University
Improving Proficiency and Efficiency of Android GUI Agents via Self-Generating Tool Actions
Sep 6, 2026Juyong Lee, Woogyeol Jin, Kimin LeeLLM Agent Tool Generation Mobile GUI Agents
KAIST
AllocBench: Measuring Online Tool Allocation Capability in LLM Agents
Jul 25, 2026Daniel Wang, Andrew XuTool-Augmented Language Model Agents AI Coding Agents
Yale University
DynamicMCPBench: A Trace-Grounded, Effect-Scored Benchmark for LLM Agents over Live MCP Servers
Jul 10, 2026Jerzy Kamiński, Ilya Galyukshev, Artem Kuznetsov +4Computer-Use Agent Benchmarks LLM Agent Evaluation
ITMO University · Central University
Tool-Making and Self-Evolving LLM Agents in Low-Latency Systems
Jul 9, 2026Kalle Kujanpää, Ning Liu, Shahnawaz Alam +4LLM Inference Efficiency LLM Agent Self-Improvement
Amazon, Fulfillment Technologies & Robotics
CacheRL:Multi-Turn Tool-Calling Agents via Cached Rollouts and Hybrid Reward
Jun 12, 2026Md Amirul Islam, Sumiran Thakur, Huancheng Chen +3LLM Agent Training Tool-Using Language Agents
Center for Advanced AI, Accenture
El Agente Forjador: Task-Driven Agent Generation for Quantum Simulation
Apr 16, 2026Zijian Zhang, Aiwei Yin, Amaan Baweja +4Multi-Agent LLM Systems AI for Science
Department of Computer Science, University of Toronto, 40 St George St., Toronto, ON M5S 2E4, Canada · Vector Institute for Artificial Intelligence, W1140-108 College St., Schwartz Reisman Innovation2026 Campus, Toronto, ON M5G 0C6, Canada · NVIDIA, 431 King St W #6th, Toronto, ON M5V 1K4, Canada +5
OpenEarth-Agent: From Tool Calling to Tool Creation for Open-Environment Earth Observation
Mar 23, 2026Sijie Zhao, Feng Liu, Xueliang Zhang +11LLM Agent Tool Generation LLM Agent Planning
Nanjing University · Shanghai Artificial Intelligence Laboratory · Shanghai Jiao Tong University +1
Separating Intelligence from Execution: A Workflow Engine for the Model Context Protocol
Mar 13, 2026Abhinav Singh ParmarLLM Agent Tool Generation LLM Agent Orchestration
A Preprint
Toward Scalable Verifiable Reward: Proxy State-Based Evaluation for Multi-turn Tool-Calling LLM Agents
Feb 18, 2026Yun-Shiuan Chuang, Chaitanya Kulkarni, Alec Chiu +8LLM Agent Tool Generation LLM Agent Evaluation
PayPal AI

Oct 5, 2026J/K move · Enter open · S save

SchemaFill: Efficient LLM Tool Calling via Slot-Parallel Speculative Decoding

Zhi-Kai Chen, Song-Yan Li, De-Chuan Zhan, Han-Jia Ye

School of Artificial Intelligence, Nanjing University, China · National Key Laboratory for Novel Software Technology, Nanjing University, China · Nanjing University, China

Speculative Decoding LLM Inference Acceleration LLM Tool Use Tool Use LLM Agent Tool Generation

Open paper AskPDF arXiv

LLM agents interact with external systems by generating structured tool calls. Given a user request, conversational context, and a catalog of tool schemas, a tool-calling model must select tools and generate their arguments, potentially producing multiple calls in a single response. Standard autoregressive decoding generates these calls token by token, incurring substantial latency for requests involving multiple calls or many argument fields. The explicit argument structure offers opportunities for parallel generation, but later argument values may depend on preceding fields and calls, so independently generated values can differ from the target model's output. We present SchemaFill, a framework for efficient LLM tool calling through slot-parallel speculative decoding. SchemaFill generates future slot values concurrently as candidates, without requiring advance knowledge of the actual call sequence or argument values. Candidates spanning multiple fields and calls are concatenated for verification by the target model under the actual output prefix. Only verified tokens are committed, and the target supplies corrections when candidates disagree. This applies target verification while exploiting parallelism across slots and calls. On Glaive and BFCL, SchemaFill achieves up to a 4.05

\times

improvement in end-to-end throughput over autoregressive decoding. Code is available at https://github.com/Czzzk/SchemaFill.