the question with id=28811 has some coding error:
極道の紋章 lastest movie database
( argmax ( lambda ?x ( mso:film.film.series ?x 極道の紋章 ) ) ( lambda ?x ( lambda ?y ( mso:film.film.initial_release_date ?x ?y ) ) ) 1 )
極道の紋章 (entity) [0,0] ||| 1 (value) [-1,-1]
superlative

the same in id=29949

Thanks for your work. But If any detail plan of releasing the KB snippets of Satori.

sorry, i donot know how to use this dataset. where is the training model? how to train the model with this dataset?

I have found that all the data of type "multi-turn-predicate" have a same error.
For example, for id=38753
who design sankt annæ plads ||| how about for florida baptist building ?
( lambda ?x ( mso:architecture.structure.architect sankt_annæ_plads ?x ) ||| ( lambda ?x ( mso:architecture.structure.architect florida_baptist_building ?x ) )
sankt_annæ_plads (entity) [2,4] @q1 ||| florida_baptist_building (entity) [3,5] @q2
multi-turn-predicate
The first sentence's logical form is missing one ')' at the end.
This error happens in every multi-turn-predicate data.

In the trainning data file, there's an error data in line 249698 as the content of this line is :

<ques t [1,1] ion en t [1,1] i t [1,1] y id=49940> t (entity) [1,1]

which should be:

<parameters id=49940> t (entity) [1,1]

请问有没有开源的能直接 解析 lambda 标注的相关工具，把lambda 的string 解析成一个结构化的对象。