Publications

A collection of my research work.

Scaling LLM Knowledge Boundaries via Distribution-Optimized Synthesis

Scaling LLM Knowledge Boundaries via Distribution-Optimized Synthesis

Songze Li, Yarong Lan, Zhongpu Bo, Zhaoyang Wang, Zhiqiang Liu, Yuan Yuan, Chengtao Gan, Menghao Qian, Enpei Niu, Xiaoke Guo, Yuanxiang Liu, Zhaoyan Gong, Xiangjin Hu, Liangyurui Liu, Jingdian Lu, Lei Liang, Jun Zhou, Huajun Chen, Wen Zhang†

EMNLP 2026 Submission 2026

We propose Knowledge Distribution-Optimized Synthesis (KDoS), a synthetic data framework that controls knowledge distribution, and find that an optimal knowledge distribution consistently exists across model sizes, data scales, and backbone architectures, maximizing the efficiency of scaling up LLM knowledge boundaries.

Last Layer Logits to Logic: Empowering LLMs with Logic-Consistent Structured Knowledge Reasoning

Last Layer Logits to Logic: Empowering LLMs with Logic-Consistent Structured Knowledge Reasoning

Songze Li, Zhiqiang Liu, Zhaoyan Gong, Xiaoke Guo, Zhengke Gui, Huajun Chen, Wen Zhang†

EMNLP 2026 Submission 2026

This paper presents Logits-to-Logic, a novel output-perspective approach addressing the Logic Drift of large language models (LLMs) in structured knowledge reasoning, with deeper insights into maintaining logic-consistent reasoning in LLMs.

ASTRA: Adaptive Structured Tree Reasoning Architecture for Complex Table Question Answering

ASTRA: Adaptive Structured Tree Reasoning Architecture for Complex Table Question Answering

Xiaoke Guo, Songze Li, Zhiqiang Liu, Zhaoyan Gong, Yuanxiang Liu, Huajun Chen, Wen Zhang†

ACL 2026 Main 2026

We introduce ASTRA, a framework that reconstructs complex tables into Logical Semantic Trees, unlocking a dual-mode reasoning capability that achieves state-of-the-art performance by effectively capturing explicit hierarchies and semantic contexts.

What's Missing in Screen-to-Action? Towards a UI-in-the-Loop Paradigm for Multimodal GUI Reasoning

What's Missing in Screen-to-Action? Towards a UI-in-the-Loop Paradigm for Multimodal GUI Reasoning

Songze Li, Xiaoke Guo, Tianqi Liu, Biao Yi, Zhaoyan Gong, Zhiqiang Liu, Huajun Chen, Wen Zhang†

ACL 2026 Findings 2026

We propose UI‑in‑the‑Loop, which reformulates GUI reasoning from Screen‑to‑Action into a cyclic Screen-UI Elements-Action paradigm, and introduce the UI Comprehension task with three metrics and the 26K UI Comprehension-Bench benchmark.

Temp-R1: A Unified Autonomous Agent for Complex Temporal KGQA via Reverse Curriculum Reinforcement Learning

Temp-R1: A Unified Autonomous Agent for Complex Temporal KGQA via Reverse Curriculum Reinforcement Learning

Zhaoyan Gong, Zhiqiang Liu, Songze Li, Xiaoke Guo, Yuanxiang Liu, Xinle Deng, Zhizhen Liu, Lei Liang, Huajun Chen, Wen Zhang

ACL 2026 Main 2026

CoG: Controllable Graph Reasoning via Relational Blueprints and Failure-Aware Refinement over Knowledge Graphs

CoG: Controllable Graph Reasoning via Relational Blueprints and Failure-Aware Refinement over Knowledge Graphs

Yuanxiang Liu, Songze Li, Xiaoke Guo, Zhaoyan Gong, Qifei Zhang, Huajun Chen, Wen Zhang†

ACL 2026 Main 2026

Enrich-on-Graph: Query-Graph Alignment for Complex Reasoning with LLM Enriching

Enrich-on-Graph: Query-Graph Alignment for Complex Reasoning with LLM Enriching

Songze Li, Zhiqiang Liu, Zhengke Gui, Huajun Chen, Wen Zhang†

EMNLP 2025 Main 2025

We propose Enrich-on-Graph (EoG), a flexible framework that leverages LLMs' prior knowledge to enrich knowledge graphs and bridge the semantic gap between structured KGs and unstructured queries for complex KGQA reasoning.

DOI