Saved in:
Bibliographic Details
Main Authors: Shi, Qiming, Kang, Zhaolu, Zhou, Yunfan, Weng, Di, Wu, Yingcai
Format: Preprint
Published: 2026
Subjects:
Online Access:https://arxiv.org/abs/2606.00593
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866913175654367232
author Shi, Qiming
Kang, Zhaolu
Zhou, Yunfan
Weng, Di
Wu, Yingcai
author_facet Shi, Qiming
Kang, Zhaolu
Zhou, Yunfan
Weng, Di
Wu, Yingcai
contents Large language models are increasingly deployed as tool-augmented agents to acquire information beyond parametric knowledge. While recent work has improved long-horizon tool-use reasoning, most approaches focus on tasks with a single correct answer. In contrast, many real-world queries require discovering a comprehensive set of valid answers, a setting known as Multi-Answer QA. This setting raises two challenges: fine-grained credit assignment over long search trajectories and reward alignment for sustained exploration beyond easy high-frequency entities. We propose SPADER, a reinforcement learning framework for long-horizon tool use in Multi-Answer QA. SPADER includes Step-wise Peer Advantage (SPA), a critic-free step-level credit assignment mechanism that aligns parallel trajectories by decision step and estimates advantages from peer returns. It also includes a diversity-aware exploration reward that promotes long-tail entity discovery by upweighting rare findings and downweighting redundant ones. Experiments on QAMPARI, Mintaka, WebQSP, and QUEST show that SPADER generally improves recall and overall F1 over prompting-based agents, outcome-supervised RL methods, and recent step-level supervision approaches. Our code and model weights are available at https://github.com/KhanCold/spader.
format Preprint
id arxiv_https___arxiv_org_abs_2606_00593
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle SPADER: Step-wise Peer Advantage with Diversity-Aware Exploration Rewards for Multi-Answer Question Answering
Shi, Qiming
Kang, Zhaolu
Zhou, Yunfan
Weng, Di
Wu, Yingcai
Computation and Language
Artificial Intelligence
Large language models are increasingly deployed as tool-augmented agents to acquire information beyond parametric knowledge. While recent work has improved long-horizon tool-use reasoning, most approaches focus on tasks with a single correct answer. In contrast, many real-world queries require discovering a comprehensive set of valid answers, a setting known as Multi-Answer QA. This setting raises two challenges: fine-grained credit assignment over long search trajectories and reward alignment for sustained exploration beyond easy high-frequency entities. We propose SPADER, a reinforcement learning framework for long-horizon tool use in Multi-Answer QA. SPADER includes Step-wise Peer Advantage (SPA), a critic-free step-level credit assignment mechanism that aligns parallel trajectories by decision step and estimates advantages from peer returns. It also includes a diversity-aware exploration reward that promotes long-tail entity discovery by upweighting rare findings and downweighting redundant ones. Experiments on QAMPARI, Mintaka, WebQSP, and QUEST show that SPADER generally improves recall and overall F1 over prompting-based agents, outcome-supervised RL methods, and recent step-level supervision approaches. Our code and model weights are available at https://github.com/KhanCold/spader.
title SPADER: Step-wise Peer Advantage with Diversity-Aware Exploration Rewards for Multi-Answer Question Answering
topic Computation and Language
Artificial Intelligence
url https://arxiv.org/abs/2606.00593