Issues

• [Bug] Under DDP, the effective lambda_weight grows with the number of GPUs

  voidism/DiffCSE#29 · Filter by author gss10282023 opened on Sep 26, 2026

• [Bug] pad_to_max_length silently skips the RTD loss

  voidism/DiffCSE#27 · Filter by author gss10282023 opened on Sep 26, 2026

• [Bug] Padding positions count as negatives in the replaced-token detection loss

  voidism/DiffCSE#25 · Filter by author gss10282025 opened on Sep 17, 2026

• 关于large版本

  voidism/DiffCSE#24 · Filter by author akakakaadsad opened on Feb 24, 2025

• “Why don’t you have a code file to convert the model format to Hugging Face format? Can your model be evaluated directly after being saved without conversion?

  voidism/DiffCSE#22 · Filter by author fzxxg opened on Nov 26, 2023

• Discrepancies in DiffSCE Code Execution and Reported Results: Seeking Insight

  voidism/DiffCSE#21 · Filter by author jasl1 opened on Sep 21, 2023

• Index out of range in self

  voidism/DiffCSE#20 · Filter by author Old-Lan opened on Jul 12, 2023

• when will the parameter 'neg_sim' be 'nan'?

  voidism/DiffCSE#19 · Filter by author N-Kingsley opened on Jun 8, 2023

• alignment & uniformity 原始碼與訓練細節

  voidism/DiffCSE#18 · Filter by author Rachel-Yeah-Lee opened on May 9, 2023

• grid serch

  voidism/DiffCSE#15 · Filter by author Linda230 opened on Feb 27, 2023

• Command to replicate transfer results

  voidism/DiffCSE#11 · Filter by author luffycodes opened on Sep 8, 2022

• RuntimeError: Input tensor at index 3 has invalid shape [14, 14], but expected [14, 17]

  voidism/DiffCSE#6 · Filter by author LZH-0225 opened on Jun 1, 2022
  2

• How to install?

  voidism/DiffCSE#2 · Filter by author stephenleo opened on Apr 29, 2022
  5