Logit Separability-Driven Samples and Multiple Class-Related Words Selection for Advancing In-Context Learning

Documenti analoghi

• Unveiling and Manipulating Prompt Influence in Large Language Models
  di: Feng, Zijian, et al.
  Pubblicazione: (2024)
• UniBias: Unveiling and Mitigating LLM Bias through Internal Attention and FFN Manipulation
  di: Zhou, Hanzhang, et al.
  Pubblicazione: (2024)
• LLMs Learn Task Heuristics from Demonstrations: A Heuristic-Driven Prompting Strategy for Document-Level Event Argument Extraction
  di: Zhou, Hanzhang, et al.
  Pubblicazione: (2023)
• Beyond the Next Token: Towards Prompt-Robust Zero-Shot Classification via Efficient Multi-Token Prediction
  di: Qian, Junlang, et al.
  Pubblicazione: (2025)
• FreeCtrl: Constructing Control Centers with Feedforward Layers for Learning-Free Controllable Text Generation
  di: Feng, Zijian, et al.
  Pubblicazione: (2024)