Teaching agents to predict and pre-execute their next tool call
11 hours ago
- LLM agents waste time waiting for tool call results, and tool-call speculation can hide this latency by predicting and pre-executing the next call.
- Existing speculators are separate models or cached traces that are poorly aligned with the agent's behavior, creating a speculator-agent gap.
- The paper proposes a self-speculating agent that unifies agent and speculator in one model, reusing prefix KV cache for efficiency.
- A joint agent-speculator reinforcement learning method is introduced, using the agent's own rollouts as speculation targets and alternating updates.
- The method improves next tool-call Hit@1 significantly (e.g., from 44.1 to 61.2 for Qwen3-4B) while preserving agent task success in search QA and conversational tool-use tasks.