What does this PR do?

Fixes #235

When PromptCompressor is created with use_slingua=True, calling compress_prompt() raised:

TypeError: XLMRobertaForTokenClassification.forward() got an unexpected keyword argument 'past_key_values'

Root cause: compress_prompt() only routed to compress_prompt_llmlingua2() when self.use_llmlingua2 was True. __init__ already had the correct if use_llmlingua2 or use_slingua: guard (line 92) to call init_llmlingua2, but the dispatch in compress_prompt was missing the or self.use_slingua half.

Fix: one-line change at line 536 of llmlingua/prompt_compressor.py:

# Before
if self.use_llmlingua2:
# After
if self.use_llmlingua2 or self.use_slingua:

Before submitting

• This PR fixes a typo or improves the docs (you can dismiss the other checks if that's the case).
• Was this discussed/approved via a Github issue? Please add a link to it if that's the case.
  Issue: [Bug]: When calling slingua through PromptCompressor, use_llmlingua2 should also be set to true. #235
• Did you make sure to update the documentation with your changes? (not applicable — no new API)
• Did you write any new necessary tests?
  Added tests/test_slingua_routing.py with 3 unit tests covering the fixed routing, existing llmlingua2 routing, and the unchanged neither-flag path. Tests use sys.modules stubs so they run without ML dependencies.

Who can review?

@iofu728 @QianhuiWu @XufangLuo @mydmdm

Generated by Claude Code
Vibe coded by ousamabenyounes