Skip to content
@CHATS-lab

CHATS-lab

Conversation, Human-AI Technology, and Safety Lab

Pinned Loading

  1. verbalized-sampling verbalized-sampling Public

    Verbalized Sampling, a training-free prompting strategy to mitigate mode collapse in LLMs by requesting responses with probabilities. Achieves 2-3x diversity improvement while maintaining quality. …

    Python 702 82

  2. persuasive_jailbreaker persuasive_jailbreaker Public

    Persuasive Jailbreaker: we can persuade LLMs to jailbreak them!

    HTML 350 29

Repositories

Showing 5 of 5 repositories
  • ToolShield Public

    Official implementation for paper "Unsafer in Many Turns: Benchmarking and Defending Multi-Turn Safety Risks in Tool-Using Agents"

    CHATS-lab/ToolShield’s past year of commit activity
    Python 8 MIT 0 0 0 Updated Feb 24, 2026
  • verbalized-sampling Public

    Verbalized Sampling, a training-free prompting strategy to mitigate mode collapse in LLMs by requesting responses with probabilities. Achieves 2-3x diversity improvement while maintaining quality. Model-agnostic framework with CLI/API for creative writing, synthetic data generation, and dialogue simulation.

    CHATS-lab/verbalized-sampling’s past year of commit activity
    Python 702 82 0 0 Updated Jan 3, 2026
  • LLMs_Encode_Harmfulness_Refusal_Separately Public
    CHATS-lab/LLMs_Encode_Harmfulness_Refusal_Separately’s past year of commit activity
    Python 29 1 1 0 Updated Dec 14, 2025
  • persuasive_jailbreaker Public

    Persuasive Jailbreaker: we can persuade LLMs to jailbreak them!

    CHATS-lab/persuasive_jailbreaker’s past year of commit activity
    HTML 350 Apache-2.0 29 1 0 Updated Oct 17, 2025
  • KokoMind Public

    KokoMind: Can LLMs Understand Social Interactions?

    CHATS-lab/KokoMind’s past year of commit activity
    JavaScript 104 Apache-2.0 8 3 0 Updated Oct 3, 2023

People

This organization has no public members. You must be a member to see who’s a part of this organization.

Most used topics

Loading…