Biblio Discovery — Demo Discovery Service

5 results for "Tu, Songjun"

Showing 1–5 · 1.15 seconds · RSS

  1. Preprint Full text

    Dynamic Dual-Granularity Skill Bank for Agentic RL

    Tu, Songjun; Xu, Chengdong; Zhang, Qichao et al.

    arXiv · 2026 · English · arXiv

    Agentic reinforcement learning (RL) can benefit substantially from reusable experience, yet existing skill-based methods mainly extract trajectory-level guidance and often lack principled mechanisms for maintaining an evolving skill memory. We propose D2Skill, a dynamic dual-granularity skill bank…

  2. Preprint Full text

    Saliency-Guided Representation with Consistency Policy Learning for Visual Unsupervised Reinforcement Learning

    Sun, Jingbo; Zhang, Qichao; Tu, Songjun et al.

    arXiv · 2026 · English · arXiv

    Zero-shot unsupervised reinforcement learning (URL) offers a promising direction for building generalist agents capable of generalizing to unseen tasks without additional supervision. Among existing approaches, successor representations (SR) have emerged as a prominent paradigm due to their…

  3. Preprint Full text

    PaperAudit-Bench: Benchmarking Error Detection in Research Papers for Critical Automated Peer Review

    Tu, Songjun; Ma, Yiwen; Lin, Jiahao et al.

    arXiv · 2026 · English · arXiv

    Large language models can generate fluent peer reviews, yet their assessments often lack sufficient critical rigor when substantive issues are subtle and distributed across a paper. In this paper, we introduce PaperAudit-Bench, which consists of two components: (1) PaperAudit-Dataset, an error…

  4. Preprint Full text

    AutoSearch: Adaptive Search Depth for Efficient Agentic RAG via Reinforcement Learning

    Sun, Jingbo; Chong, Wenyue; Tu, Songjun et al.

    arXiv · 2026 · English · arXiv

    Agentic retrieval-augmented generation (RAG) systems enable large language models (LLMs) to solve complex tasks through multi-step interaction with external retrieval tools. However, such multi-step interaction often involves redundant search steps, incurring substantial computational cost and…

  5. Preprint Full text

    $π$-Play: Multi-Agent Self-Play via Privileged Self-Distillation without External Data

    Zhang, Yaocheng; Zhu, Yuanheng; Chong, Wenyue et al.

    arXiv · 2026 · English · arXiv

    Deep search agents have emerged as a promising paradigm for addressing complex information-seeking tasks, but their training remains challenging due to sparse rewards, weak credit assignment, and limited labeled data. Self-play offers a scalable route to reduce data dependence, but conventional…