Now we should see how the Shi'as have solved the problem of such ahadith.
The Shi'as do not believe in the immunity of any writer, commentator or narrator from mistakes, and, therefore, they do not take any collection of hadith to be completely valid and correct. The only book which is completely immune from any mistake is the Qur'an.
There are four early collections of Shi'a ahadith which are together called "al-Kutub al-Arb'ah" (the four books): al-Kafi of Kulayni, Man la Yahdhuruhu' l-Faqih of Saduq, Tahdhibu 'l-Ahkam and al-Istibsar of Tusi. Although these books are held in great esteem, the Shi'as have never called them "sihah". Consequently, they are not fettered by any hadith written there simply because it is in one of the four books. Instead, they subject all ahadith in all these books to strict tests, as to their narrators (asnad) and dirayah, and examine whether a given hadith conforms with the Qur'an, the accepted sayings of the Ma'sumin and the known facts. If a hadith passes these tough test, then it is accepted. If not, then it is re-interpreted in an acceptable way, failing which it is rejected outright.
It should be mentioned here that an overwhelming part of traditions concerning tahrif is defective and weak as far as their chains of narrators are concerned. Even then, some of those traditions maybe taken to denote that there has occurred misinterpretation in some verses, giving them wrong meaning. Another group of traditions may easily be construed to mention marginal explanatory notes of the reciters.
But there still remain many traditions which cannot be explained in either way. And our scholars unhesitatingly have rejected them because they go against the Qur'an and the sunnah, and are contrary to the ijma' of the ummah that there has never been any addition in or omission from the Qur'an.
Muhaqqiq al-Kalbasi has said, "All these reports which speak of tahrif are against the ijma' of the ummah (with exception of a few insignificant persons)." 
The commentator of al-Wafiyah, Muhaqqiq al-Baghdadi has clearly stated, by quoting from Muhaqqiq al-Karaki (who has written a complete tract on the subject) that: "The traditions which speak of omission must either be reinterpreted or rejected. Any tradition which is contradictory to the Qur'an, the acknowledged sunnah and the ijma' must be discarded if it has no room for interpretation or justifiable explanation." 
A tradition, recorded in al-Kafi is quoted here to give an example in practice of what we mean when we speak of reinterpretation or justifiable explanation: Abu 'Abdillah [al-Sadiq] (peace be upon him) said, "The Qur'an which was brought by Jibrail (peace be upon him) to Muhammad (Mercy of Allah be on him and his progeny) is seventeen thousand verses." 
Shaykh as-Saduq has written in his Kitabu'l-Itiqaddat, what in my view amounts to a reinterpretation of this hadith. He writes, "We say that so much of revelation has come down, which is not a part of the Qur'an, that if it were to be collected, its extent would undoubtedly be 17,000 verses. And this, for example, is like the saying of Gabriel (a.s.) to the Prophet (peace be upon him & his progeny): 'Allah says to thee, O Muhammad, deal gently with My creatures, in the same manner as I do."'
He goes on quoting many such ahadith qudsiyah until he concludes by saying: "There are many such (ahadith qudsiyah) all of which are revelations, but do not form part of the Qur'an. If they had been (part of the Qur'an), they would surely have been included in it, and not excluded from it." 
If one is not prepared to accept this explanation because the tradition speaks about "the Qur'an", then we will discard this hadith without hesitation. Although the number (17,000) given in this tradition is much smaller than the one given by the 2nd Caliph (one million and twenty seven thousand letters in the Qur'an),  it is never-the-less three times bigger than the actual number of the verses (which is some six thousand two hundred and thirty six verses). 
This is the Shi'a method of resolving the problem of the ahadith
on tahrif. Anyone studying it with unbiased mind and heart would
readily accept it as the only correct solution for such ahadith
because it is based on the method formulated by the Prophet (peace be upon
him & his progeny) and expressed by Imam Ja'far as-Sadiq (a.s.) as
follows: "...And so whatever conforms to the Book of Allah, take it; and
whatever is contrary to the Book of Allah, leave it .. " 
Quoted in al Bayan, p. 253
al Kulayni, al Kafi, vol. 2 (Tehran: a; Matba'a al Islamiyya, 1388),
Saduq, Kitabu 'l-Itiqadat, p. 63-65
As-Suyuti, al-Itqan, vol. 2, p. 70
See Mahmud Ruhani, al Mu'jamu 'l Ihsai (Mashad: 1990) p. 168. Those
who do note take the trouble of counting, go on writing that there are
6666 verses in the Qur'an. It is one more example of Muslims tragic carelessness
towards the Qur'an.
al Hurr al Amili, Wasailu 'sh Shiah, vol. 3 (Kitabu 'l qadha: bab
wujuhi 'l jam bayna 'l ahadithi 'l mukhtalifah), p. 380
 Died in 1994.