Skip to content

Commit 60a759b

Browse files
authored
Merge pull request #739 from openvanilla/dev/update-heterophony-instructions
Discourage agents from suggesting changes to `heterophony{1,2,3}.list`
2 parents 489c893 + 5b03987 commit 60a759b

1 file changed

Lines changed: 3 additions & 11 deletions

File tree

Source/Data/AGENTS.md

Lines changed: 3 additions & 11 deletions
Original file line numberDiff line numberDiff line change
@@ -126,19 +126,11 @@ Format: `character bopomofo pinyin tone tag`
126126

127127
### Handling Heterophony Characters (破音字)
128128

129-
For characters with multiple pronunciations (e.g., 行: ㄏㄤˊ vs ㄒㄧㄥˊ):
129+
Do NOT suggest any changes to `heterophony1.list`, `heterophony2.list` or `heterophony3.list`. If a PR contains any changes to those files, highlight them and ask human reviewers to pay attention to them.
130130

131-
1. Place most common reading in `heterophony1.list`, less common in `heterophony2.list`, rare in `heterophony3.list`
132-
2. Add both readings to `BPMFMappings.txt`
131+
For Mandarin heteronyms, an entry in `heterophony1.list` sets the primary reading and causes other readings of the same character to be demoted. This can be problematic for high-frequency characters whose other readings are also common. To compensate, `heterophony2.list` and `heterophony3.list` "promote" those readings back, but with a frequency discount to ensure they don't outrank the primary one.
133132

134-
Example:
135-
```bash
136-
# heterophony1.list
137-
行 ㄏㄤˊ
138-
139-
# heterophony2.list
140-
行 ㄒㄧㄥˊ
141-
```
133+
This is a complex language model tuning mechanism, so agents must NOT suggest changes on their own.
142134

143135
### Adding Emojis and Symbols
144136

0 commit comments

Comments
 (0)