LLM Agents Already Know When to Call Tools -- Even Without Reasoning
cs.CL
Submitted: 2026-05-10
Updated: 2026-11-08
Comments: Accepted to NeurIPS 2026
Code: https://github.com/Trustworthy-ML-Lab/when2tool
License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Terminology
Sources
- $\tau^2$-Bench: Evaluating Conversational Agents in a Dual-Control Environment
- Discovering Latent Knowledge in Language Models Without Supervision
- SWE-bench: Can Language Models Resolve Real-World GitHub Issues?
- Search-R1: Training LLMs to Reason and Leverage Search Engines with Reinforcement Learning
- Language Models (Mostly) Know What They Know
- MigrationBench: Repository-Level Code Migration Benchmark from Java 8
- Terminal-Bench: Benchmarking Agents on Hard, Realistic Tasks in Command Line Interfaces
- ToolLLM: Facilitating Large Language Models to Master 16000+ Real-world APIs
- DR Tulu: Reinforcement Learning with Evolving Rubrics for Deep Research
- Concept Bottleneck Large Language Models
- Steer2Edit: From Activation Steering to Component-Level Editing
- ReflCtrl: Controlling LLM Reflection Efficiently via Representation Engineering
- Toward Efficient Agents: Memory, Tool learning, and Planning
- $\tau$-bench: A Benchmark for Tool-Agent-User Interaction in Real-World Domains
- Representation Engineering: A Top-Down Approach to AI Transparency
Related papers
- ALARM: Audio-Language Alignment for Reasoning Models
- ThinkSum: Probabilistic reasoning over sets using large language models
- Coherence boosting: When your pretrained language model is not paying enough attention
- From Facts to Insights: A Persona-Driven Dual Memory Framework and Dataset for Role-Playing Agents
- Improving General Role-Playing Agents via Psychology-Grounded Reasoning and Role-Aware Policy Optimization
- The Art of Asking: Multilingual Prompt Optimization for Synthetic Data