Skip to content
View HrxuAlbert's full-sized avatar
🎯
Focusing
🎯
Focusing

Block or report HrxuAlbert

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Maximum 250 characters. Please don’t include any personal information such as legal names or email addresses. Markdown is supported. This note will only be visible to you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
HrxuAlbert/README.md

Haoran Xu

I am a PhD student at the University of Glasgow, working on trustworthy multi-agent AI systems, selective commitment, and Byzantine-resilient semantic collaboration.

My current research studies how multi-agent AI systems can decide when to commit, when to abstain, and when to safely abort under semantic poisoning, adversarial disagreement, and unreliable evidence.

Research Interests

  • Trustworthy Multi-Agent AI
  • LLM-Agent Collaboration
  • Selective Commitment and Safe Abstention
  • Byzantine-Resilient Semantic Systems
  • Evidence-Aware Decision Protocols
  • AI Safety and Trustworthy AI Infrastructure

Current Research Direction

I am developing protocol-level mechanisms for trustworthy multi-agent collaboration, with a focus on semantic commitment, evidence-aware compatibility, certificate-based finality, and safe-abort mechanisms.

Selected Projects

  • CRSE / FBA Encoder — Robust semantic encoder for Byzantine-resilient semantic agreement.
  • LLM API Demos — Lightweight examples for structured LLM interaction and agent-related experimentation.
  • Work Planner — Local-first academic productivity tool for planning and execution.

Links

Pinned Loading

  1. FBA_ENCODER FBA_ENCODER Public

    Robust semantic encoder for Byzantine-resilient multi-agent agreement.

    Python 1

  2. HrxuAlbert.github.io HrxuAlbert.github.io Public

    Personal academic website of Haoran Xu.

    HTML

  3. LLM-API-demos LLM-API-demos Public

    Minimal demos for structured LLM API usage, agent workflows, and research prototyping.

    Python

  4. cherry-pick-override cherry-pick-override Public

    Code and data release for the paper 'Cherry-pick Override: Unsafe Directional Commitment in LLM Judges under Mixed Evidence'

    Python

  5. H-CSC H-CSC Public

    Hierarchical Certified Semantic Commitment (H-CSC): Byzantine-fault-tolerant finality control for LLM-agent collaboration

    Python