Litu Ou

Litu Ou

PhD student at the University of Edinburgh working on long-context LLMs and agents

Litu Ou

PhD Student, ILCC, University of Edinburgh

About Me

I am a first-year PhD student at the University of Edinburgh, supervised by Prof. Mirella Lapata. My research focuses on long-context understanding and long-horizon reasoning in large language models, with applications to question-answering, summarization and web agents. Previously, I interned at Tongyi Lab, where I worked on the Tongyi DeepResearch project.

Research Focus

Long-Context Understanding & Retrieval

Studying how LLMs retrieve, integrate, and reason over evidence from long inputs such as books, reports, and multi-step search traces.

Web Agents & Post-Training

Building and evaluating LLM agents that search, use tools, and solve long-horizon information-seeking tasks, with an emphasis on synthetic data, SFT, RL, and test-time control.

Context Management & Memory

Designing summarization and memory mechanisms that help agents maintain useful state under limited context budgets.

News

  1. 2026.05
    One paper accepted to SIGIR 2026.
  2. 2026.05
    BrowseConf accepted to Findings of ACL.
  3. 2026.01
    Started as a teaching assistant for Foundations of Natural Language Processing.
  4. 2025.09
    Started my PhD at the University of Edinburgh.
  5. 2025.06
    Started a research internship at Tongyi Lab.

Selected Work

2026 - Findings of ACL

BrowseConf

Confidence-guided test-time scaling for web agents.

Web AgentsTest-Time ScalingReliability

Research Experience

  1. 2025.09 - Present
    PhD Student, Institute for Language, Cognition and Computation, University of Edinburgh

    Researching long-context and long-horizon LLM systems, supervised by Prof. Mirella Lapata.

  2. 2025.06 - 2025.09
    Research Intern, Tongyi DeepResearch Team

    Worked on post-training, test-time control, and scalable RL pipelines for LLM-based web agents.

  3. 2024.08 - 2025.05
    Research Intern (Remote), Tsinghua University

    Worked on tool learning and multi-step reasoning with large language models.

  4. 2023.06 - 2024.11
    Research Assistant, University of Edinburgh

    Worked on retrieval-enhanced and hierarchical methods for faithful book-length summarization.

Publications and Preprints

  1. Revisiting Text Ranking in Deep Research
    Chuan Meng, Litu Ou, Sean MacAvaney, Jeff Dalton - SIGIR, 2026
  2. Uncertainty-Aware Gradient Signal-to-Noise Data Selection for Instruction Tuning
    Zhihang Yuan, Chengyu Yue, Long Huang, Litu Ou, Lei Shi - arXiv, 2026
  3. Tongyi DeepResearch Technical Report
    Tongyi DeepResearch Team, including Litu Ou - arXiv, 2025
  4. BrowseConf: Confidence-Guided Test-Time Scaling for Web Agents
    Litu Ou, Kuan Li, Huifeng Yin, Liwen Zhang, Zhongwang Zhang, Xixi Wu, Rui Ye, Zile Qiao, Yong Jiang, Pengjun Xie, Fei Huang, Jingren Zhou - Findings of ACL, 2026
  5. WebSailor-V2: Bridging the Chasm to Proprietary Agents via Synthetic Data and Scalable Reinforcement Learning
    Kuan Li, Zhongwang Zhang, Huifeng Yin, Rui Ye, Yida Zhao, Liwen Zhang, Litu Ou, et al. - ICLR, 2026
  6. ReSum: Unlocking Long-Horizon Search Intelligence via Context Summarization
    Xixi Wu, Kuan Li, Yida Zhao, Liwen Zhang, Litu Ou, Huifeng Yin, Zhongwang Zhang, et al. - arXiv, 2025
  7. WebSailor: Navigating Super-human Reasoning for Web Agent
    Kuan Li, Zhongwang Zhang, Huifeng Yin, Liwen Zhang, Litu Ou, Jialong Wu, et al. - arXiv, 2025
  8. ToLeaP: Rethinking Development of Tool Learning with Large Language Models
    Haotian Chen, Zijun Song, Boye Niu, Ke Zhang, Litu Ou, Yaxi Lu, Zhong Zhang, Xin Cong, Yankai Lin, Zhiyuan Liu, Maosong Sun - arXiv, 2025
  9. Context-Aware Hierarchical Merging for Long Document Summarization
    Litu Ou and Mirella Lapata - Findings of ACL, 2025
  10. Open-Source vs Close-Source: The Context Utilization Challenge
    Litu Ou - ICLR Blogpost Track, 2025
  11. Chain-of-Thought Hub: A Continuous Effort to Measure Large Language Models' Reasoning Performance
    Yao Fu, Litu Ou, Mingyu Chen, Yuhao Wan, Hao Peng, Tushar Khot - ICML, 2023
  12. Specializing Smaller Language Models towards Multi-Step Reasoning
    Yao Fu, Hao Peng, Litu Ou, Ashish Sabharwal, Tushar Khot - ICML Oral/Poster, 2023

Academic Service / Teaching

  1. 2026.02
    Reviewer, LLA Workshop at ICLR
  2. 2026.01 - 2026.05
    Teaching Assistant, Foundations of Natural Language Processing, University of Edinburgh