Speculate While You Reason: Teaching Agents to Predict Their Next Tool Call via Joint Agent-Speculator RL

• URL: http://arxiv.org/abs/2607.25816v1
• Date: Tue, 28 Jul 2026 15:00:10 GMT
• Title: Speculate While You Reason: Teaching Agents to Predict Their Next Tool Call via Joint Agent-Speculator RL
• Abstract summary: Self-speculating agent solves tasks in agent mode and predicts its next tool call from partial trajectories in speculator mode.<n>Joint agent-speculator reinforcement learning method derives speculation targets from the agent's own rollouts and alternates agent and speculator updates.
• Score: 26.192522057926737
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Large language model agents often spend substantial wall-clock time waiting for tool call results. Tool-call speculation can hide this latency by predicting and pre-executing an agent's next tool call if the prediction matches the agent's eventual tool call, but existing speculators are typically separate draft models or cached traces that are poorly aligned with the deployed agent's own behavior. We identify this speculator-agent gap and show that the target agent itself is a strong next-call speculator. This points to a simpler design: unifying the agent and speculator within the same model. In this paper, we introduce the self-speculating agent, a single model that both solves tasks in agent mode and predicts its next tool call from partial trajectories in speculator mode, fully reusing prefix KV cache. To enable this dual-mode agent without degrading performance, we propose a joint agent-speculator reinforcement learning method, which derives speculation targets from the agent's own rollouts and alternates agent and speculator updates. Across agentic search QA and conversational tool-use agentic tasks, our method improves average next tool-call Hit@1 from 44.1 to 61.2 for Qwen3-4B and from 48.9 to 66.3 for Qwen3.5-4B, while preserving agent task success.

Related papers

• AgentFugue: Agent Scaling for Long-Horizon Tasks through Collective Reasoning [57.72181998599248]
  AgentFugue is a collective reasoning framework built around a shared reasoning hub.<n>As peer agents explore the same task in parallel, the hub records concise notes on what each agent has established, attempted, or ruled out.<n>Our results suggest that collective reasoning can turn scaling out peer agent systems into a distinct source of capability gains.
  arXiv  Detail & Related papers  (2026-05-23T09:21:51Z)
• Speculative Interaction Agents: Building Real-Time Agents with Asynchronous I/O and Speculative Tool Calling [64.40340291543971]
  We propose Speculative Interaction Agents to enable real-time interaction with small edge-scale models.<n>We also present a clock-based training methodology that adapts the model to handle streaming inputs and asynchronous responses.<n>This approach provides 1.6-2.2$times$ speedups with the Qwen2.5-3B-Instruct and Llama-3.2-3B-Instruct models across multiple tool calling benchmarks.
  arXiv  Detail & Related papers  (2026-05-13T11:20:52Z)
• Act While Thinking: Accelerating LLM Agents via Pattern-Aware Speculative Tool Execution [18.50317396128381]
  PASTE is a Pattern-Aware Speculative Tool Execution method designed to hide tool latency through speculation.<n>We show that PASTE reduces average task completion time by 48.5% and improves tool execution throughput by 1.8x.
  arXiv  Detail & Related papers  (2026-03-19T13:36:50Z)
• Behind the Prompt: The Agent-User Problem in Information Retrieval [4.563318916484434]
  User models in information retrieval rest on a foundational assumption that observed behavior reveals intent.<n>For any action an agent takes, a hidden instruction could have produced identical output - making intent non-identifiable at the individual level.<n>We investigate the agent-user problem through a large-scale corpus from an agent-native social platform.
  arXiv  Detail & Related papers  (2026-03-04T01:42:14Z)
• AgentSentry: Mitigating Indirect Prompt Injection in LLM Agents via Temporal Causal Diagnostics and Context Purification [25.817251923574286]
  We propose a novel inference-time detection and mitigation framework for large language model (LLM) agents.<n>AgentSentry is the first inference-time defense to model multi-turn IPI as a temporal causal takeover.<n>We evaluate AgentSentry on the textscAgentDojo benchmark across four task suites, three IPI attack families, and multiple black-box LLMs.
  arXiv  Detail & Related papers  (2026-02-26T07:59:10Z)
• AgentArk: Distilling Multi-Agent Intelligence into a Single LLM Agent [57.10083973844841]
  AgentArk is a novel framework to distill multi-agent dynamics into the weights of a single model.<n>We investigate three hierarchical distillation strategies across various models, tasks, scaling, and scenarios.<n>By shifting the burden of computation from inference to training, the distilled models preserve the efficiency of one agent while exhibiting strong reasoning and self-correction performance of multiple agents.
  arXiv  Detail & Related papers  (2026-02-03T19:18:28Z)
• BackdoorAgent: A Unified Framework for Backdoor Attacks on LLM-based Agents [58.83028403414688]
  Large language model (LLM) agents execute tasks through multi-step workflow that combine planning, memory, and tool use.<n>Backdoor triggers injected into specific stages of an agent workflow can persist through multiple intermediate states and adversely influence downstream outputs.<n>We propose textbfBackdoorAgent, a modular and stage-aware framework that provides a unified agent-centric view of backdoor threats in LLM agents.
  arXiv  Detail & Related papers  (2026-01-08T03:49:39Z)
• SmartSnap: Proactive Evidence Seeking for Self-Verifying Agents [45.71333459905404]
  SmartSnap is a paradigm shift from passive, post-hoc verification to proactive, in-situ self-verification by the agent itself.<n>We introduce the Self-Verifying Agent, a new type of agent designed with dual missions: to complete a task and to prove its accomplishment with curated evidences.<n>Experiments on mobile tasks across model families and scales demonstrate that our SmartSnap paradigm allows training LLM-driven agents in a scalable manner.
  arXiv  Detail & Related papers  (2025-12-26T14:51:39Z)
• InfoAgent: Advancing Autonomous Information-Seeking Agents [143.15973604285304]
  We introduce InfoAgent, a deep research agent powered by an innovative data synthesis pipeline and orchestrated web search tools.<n>With our methods, InfoAgent achieves 15.3% accuracy on BrowseComp, 29.2% on BrowseComp-ZH, and 40.4% on Xbench-DS.
  arXiv  Detail & Related papers  (2025-09-29T17:59:57Z)
• Scaling Agents via Continual Pre-training [80.97989245493326]
  We propose incorporating Agentic Continual Pre-training (Agentic CPT) into the deep research agents training pipeline to build powerful agentic foundational models.<n>We evaluate our AgentFounder-30B on 10 benchmarks and achieve state-of-the-art performance while retains strong tool-use ability.
  arXiv  Detail & Related papers  (2025-09-16T17:57:19Z)
• AgentFormer: Agent-Aware Transformers for Socio-Temporal Multi-Agent Forecasting [25.151713845738335]
  We propose a new Transformer, AgentFormer, that jointly models the time and social dimensions. Based on AgentFormer, we propose a multi-agent trajectory prediction model that can attend to features of any agent at any previous timestep. Our method significantly improves the state of the art on well-established pedestrian and autonomous driving datasets.
  arXiv  Detail & Related papers  (2021-03-25T17:59:01Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.