You can exclude the wordforms you do not need from the search results. To do that, type the base form into the Lemma input box, then type the minus sign followed by the wordform you do not needed, without a space and in quotation marks. For example, searching for the forms of the verb прочить in the disambiguated part of the main corpus yileds many examples with the word прочь (as in Прочь отсюда! ) which is a homonym to the imperative form of this verb (as in Не прочь меня в начальники). To get rid of these examples, you can type прочить -"прочь" (but bear in mind that in this case the imperative forms will not be found either).
To exclude a form that you do not need you can also use the minus sign in the Wordform box, putting it before the form.
You can remove as many forms as you need. In the example with the verb прочить, you would not only need to remove the form прочь, but also the form прочен, which is also homonymous because it could be a participle of прочить or the short form of the adjective прочный: прочить -"прочь" -"прочен".
If a lemma is specified as a wildcard with an asterisk, the minus sign (without quotes) can be used to exclude unnecessary lemmas. For example, the query *вед -швед will find all the lexemes ending in -вед such as искусствовед, but not швед.
The minus operator can also be used with the grammatical features or additional tags. To do that you will need to find the name(s) of the tag(s) you do not need. You can look them up in the list or find them through the popup menu. After you found the name of the tag, you can include it into the corresponding box of the query preceded by the minus sign. The priority of the operations NOT, AND, OR and brackets is usual. Accordingly, by adding a minus sign (NOT) to the query string before the brackets, you exclude everything in brackets, by adding a minus before the token, you exclude only the nearest token.