BiLD: Bi-directional Logits Difference Loss for Large Language Model Distillation

Documenti analoghi

• Bi-directional Bias Attribution: Debiasing Large Language Models without Modifying Prompts
  di: Lin, Yujie, et al.
  Pubblicazione: (2026)
• BiTA: Bi-Directional Tuning for Lossless Acceleration in Large Language Models
  di: Lin, Feng, et al.
  Pubblicazione: (2024)
• UNDIAL: Self-Distillation with Adjusted Logits for Robust Unlearning in Large Language Models
  di: Dong, Yijiang River, et al.
  Pubblicazione: (2024)
• Bi-Chainer: Automated Large Language Models Reasoning with Bidirectional Chaining
  di: Liu, Shuqi, et al.
  Pubblicazione: (2024)
• BiMark: Unbiased Multilayer Watermarking for Large Language Models
  di: Feng, Xiaoyan, et al.
  Pubblicazione: (2025)