You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
The ambiguity has resulted in bugs at nearly every step of development, from the previous FSM class, to the original set of CFG LogitsProcessors, just from my recollection.
How would you like it to behave?
Investigate whether is_termination_accepting_state is necessary anywhere. It probably is necessary to determine whether EOS is legal in some places.
Implement is_halted_state. This is the set of states which must produce EOS
What behavior of the library made you think about the improvement?
The verbiage is a bit confusing.
RegexGuide.is_final_state
currently returns both EOS-required AND EOS-accepting states.This has resulted in a few bugs
Guide
in vLLM integration #856The ambiguity has resulted in bugs at nearly every step of development, from the previous
FSM
class, to the original set of CFG LogitsProcessors, just from my recollection.How would you like it to behave?
is_termination_accepting_state
is necessary anywhere. It probably is necessary to determine whether EOS is legal in some places.is_halted_state
. This is the set of states which must produce EOSis_final_state
s.t. they use the appropriate new function, e.g. https://github.com/outlines-dev/outlines/blob/4f8433d8d6633b0780c3a6c27981f9adffbe49f5/outlines/generate/generator.py#L94The text was updated successfully, but these errors were encountered: