fix: filter noisy AI tags

Add an editable AI tag removal list and apply it to WD/JoyTag output before tags are stored. Clean existing generated AI tags during database migration while leaving manual user tags untouched.

src-tauri/src/ai_tag_filter.rs

@@ -0,0 +1,45 @@
+/// AI-generated tags that are too broad/noisy to be useful in this gallery.
+/// Edit this list to change what the tagger removes. Manual user tags are not
+/// affected.
+pub const AI_TAG_REMOVAL_LIST: &[&str] = &["1girl", "1boy", "no humans", "2girls", "2boys"];
+
+fn normalize_tag_for_removal(tag: &str) -> String {
+    tag.trim()
+        .chars()
+        .filter(|c| !matches!(c, ' ' | '_' | '-'))
+        .flat_map(char::to_lowercase)
+        .collect()
+}
+
+pub fn is_removed_ai_tag(tag: &str) -> bool {
+    let normalized = normalize_tag_for_removal(tag);
+    AI_TAG_REMOVAL_LIST
+        .iter()
+        .any(|blocked| normalize_tag_for_removal(blocked) == normalized)
+}
+
+#[cfg(test)]
+mod tests {
+    use super::*;
+
+    #[test]
+    fn removed_ai_tags_match_common_spellings() {
+        for tag in [
+            "1girl",
+            "1 girl",
+            "1_girl",
+            "1-girl",
+            "NO_HUMANS",
+            "no humans",
+        ] {
+            assert!(is_removed_ai_tag(tag), "{tag} should be removed");
+        }
+    }
+
+    #[test]
+    fn removed_ai_tags_do_not_match_unrelated_tags() {
+        for tag in ["girl", "boy", "humans", "solo", "landscape"] {
+            assert!(!is_removed_ai_tag(tag), "{tag} should be kept");
+        }
+    }
+}

src-tauri/src/db.rs

@@ -1,4 +1,4 @@
-use crate::vector;
+use crate::{ai_tag_filter, vector};
 use anyhow::Result;
 use r2d2::Pool;
 use r2d2_sqlite::SqliteConnectionManager;
@@ -378,6 +378,42 @@ pub fn migrate(conn: &Connection) -> Result<()> {
     )?;
 
     vector::migrate(conn)?;
+    remove_filtered_ai_tags(conn)?;
+    Ok(())
+}
+
+fn remove_filtered_ai_tags(conn: &Connection) -> Result<()> {
+    let mut stmt = conn.prepare("SELECT id, tag FROM image_tags WHERE source = 'ai'")?;
+    let filtered_ids = stmt
+        .query_map([], |row| {
+            Ok((row.get::<_, i64>(0)?, row.get::<_, String>(1)?))
+        })?
+        .filter_map(|row| match row {
+            Ok((id, tag)) if ai_tag_filter::is_removed_ai_tag(&tag) => Some(Ok(id)),
+            Ok(_) => None,
+            Err(error) => Some(Err(error)),
+        })
+        .collect::<rusqlite::Result<Vec<_>>>()?;
+    drop(stmt);
+
+    if filtered_ids.is_empty() {
+        return Ok(());
+    }
+
+    let tx = conn.unchecked_transaction()?;
+    {
+        let mut delete_stmt = tx.prepare("DELETE FROM image_tags WHERE id = ?1")?;
+        for id in &filtered_ids {
+            delete_stmt.execute([id])?;
+        }
+    }
+    tx.execute("DELETE FROM tag_cloud_cache", [])?;
+    tx.commit()?;
+
+    log::info!(
+        "Removed {} filtered AI tag(s) from existing library data",
+        filtered_ids.len()
+    );
     Ok(())
 }
 

src-tauri/src/lib.rs

@@ -1,3 +1,4 @@
+mod ai_tag_filter;
 mod captioner;
 mod color;
 mod commands;

src-tauri/src/tagger.rs

@@ -1,3 +1,4 @@
+use crate::ai_tag_filter;
 use anyhow::Result;
 use hf_hub::{api::sync::Api, Repo, RepoType};
 use image::{imageops::FilterType, DynamicImage, ImageReader};
@@ -759,6 +760,7 @@ impl WdTagger {
             .filter(|(entry, prob)| {
                 (entry.category == GENERAL_CATEGORY || entry.category == CHARACTER_CATEGORY)
                     && **prob >= threshold
+                    && !ai_tag_filter::is_removed_ai_tag(&entry.name)
             })
             .map(|(entry, prob)| TagResult {
                 tag: entry.name.clone(),
@@ -1137,9 +1139,11 @@ fn joytag_tags_from_logits(
         .zip(logits.iter())
         .filter_map(|(name, logit)| {
             let confidence = sigmoid(*logit);
-            (confidence >= threshold).then(|| TagResult {
+            (confidence >= threshold && !ai_tag_filter::is_removed_ai_tag(name)).then(|| {
-                tag: name.clone(),
+                TagResult {
-                confidence,
+                    tag: name.clone(),
+                    confidence,
+                }
             })
         })
         .collect();