cs.AINov 15, 2026

Interactive Structured Environment Scaling for Large Language Model Agentic Reasoning

Authors: Peng Yu, Luoyi Fu, Xinbing Wang, Ying Wen

Organizations: Department of Computer Science and Engineering, Shanghai Jiao Tong University · School of Artificial Intelligence and John Hopcroft Center, Shanghai Jiao Tong University

Abstract

Reinforcement learning via environmental interaction is a cornerstone post-training paradigm for evolving large language models into autonomous agents capable of multi-turn tool use and exploration. Because the environment defines what the policy observes and how actions are verified, its structure directly bounds the capabilities an agent can acquire. Yet existing structured and retrieval environments rely on upstream pre-filtering---extracting a compact subgraph around the gold answer or retrieving fixed passages before the policy acts. We show that such pre-filtering introduces a fundamental measurement confound: because per-question subgraphs enclose ground-truth paths, they compress the candidate space by orders of magnitude and leak answer locations via local topology, allowing a question-blind degree heuristic to surface gold answers far above chance. To restore path selection, candidate pruning, and optimal stopping as genuine policy decisions, we propose RISE (Interactive Structured Environment), which transitions structured reinforcement learning from static subgraphs to full-scale interactive databases along two coupled axes. First, Decision-Space Scaling exposes the unpruned knowledge graph via a five-stage graph-agnostic typed intent adapter under a bounded call budget and a multiplicative grounding reward gate. Second, Learning-Frontier Scaling overcomes within-group variance collapse under group-normalized objectives by uniting model-independent structural tier progression with a self-evolving competence-frontier bandit. Across nineteen benchmarks and three model backbones, RISE substantially advances in-domain and unseen-schema knowledge-graph reasoning, transfers zero-shot to open-domain passage retrieval, and significantly enhances out-of-domain mathematical and logical deduction.

Figures & tables

Appendix figures & tables1 asset

Supplementary material from the paper’s appendix.

Appendix

Explore similar work

CardsList
  1. Better Reasoning, Not Just Bigger Models: A Structure-Grounded Framework for Efficient and Reliable Inference

    Oct 22, 2026Amirhossein Aghaei, Liangyan Li, Shiyi Yang +2LLM Reasoning with GraphsLLM Grounding

  2. GraphScout: Empowering Large Language Models with Intrinsic Exploration Ability for Agentic Graph Reasoning

    Mar 2, 2026Yuchen Ying, Weiqi Jiang, Tongya Zheng +4LLM Reasoning with GraphsAgentic RAG

  3. AgenticRag-R1: Agentic Reinforcement Learning with Stack Memory for Multi-Step Reasoning, Retrieval and Memorizing

    Aug 30, 2026Xinke Jiang, Yue Fang, Zhibang Yang +12Retrieval-Augmented GenerationAgentic RAG