docs: update inference prompts and client options to canonical instTrans format

README.md

@@ -129,8 +129,11 @@ The following settings describe the released [`translate.py`](https://github.com
 | Model ID | `IndexTeam/Index-Translate-9B`; `INDEX_MODEL` changes the default, and `--model` overrides it |
 | Source language | `auto`; override with `--source` / `-s`; auto omits the source-language name from the prompt |
 | Target language | `en`; override with `--target` / `-t` |
-| Instruction | Optional custom instruction with `--instruction` / `-i` (e.g. style, format, constraints) |
-| Glossary | Optional terminology mappings with `--glossary` / `-g` (`src:tgt` or JSON string); enforces strict term preservation |
+| Hard constraints | `--hard` / `-H` (repeatable, e.g. format preservation, code/tag protection) |
+| Soft constraints | `--soft` / `-S` (repeatable, e.g. style/tone adaptation, domain sense disambiguation) |
+| Glossary | Terminology mappings with `--glossary` / `-g` (`src:tgt` or JSON path); formatted into canonical hard term constraints |
+| Genre / Domain | Source text genre/domain with `--genre` / `-d` (default: `文本`, e.g. `专栏文章` / `学术论文` / `小说` / `结构化数据`) |
+| Instruction | General instruction with `--instruction` / `-i` (formatted into instTrans constraints) |
 | Raw prompt | `--raw-prompt`; directly send input text without template wrapping |
 | Endpoint | `http://127.0.0.1:8000/v1`; `OPENAI_BASE_URL` changes the default, and `--base-url` overrides it |
 | API key | `EMPTY` for local serving; `OPENAI_API_KEY` changes the default, and `--api-key` overrides it |
@@ -142,7 +145,7 @@ The following settings describe the released [`translate.py`](https://github.com
 
 The client maps supported short language codes to Chinese language names in the prompt. Provide the full language name for a code outside its mapping.
 
-For standard translation:
+For standard unconstrained translation:
 
 ```text
 请将以下{source-language}文本翻译为{target-language}，直接输出翻译结果，不要进行任何解释。
@@ -150,16 +153,19 @@ For standard translation:
 {source-text}
 ```
 
-When instruction or glossary constraints are provided:
+For constrained translation, the model follows the canonical **instTrans** format:
 
 ```text
-请将以下{source-language}文本翻译为{target-language}。
-翻译要求：{instruction}
-请严格遵循以下术语对应关系进行翻译：
-- {src} -> {tgt}
-直接输出翻译结果，不要进行任何解释。
+请将以下{source-language}{genre}翻译成{target-language}，并且严格遵循所有约束要求。
 
+【源文】
 {source-text}
+
+【约束要求】
+1. 【硬性要求】...
+2. 【注意】...
+
+只输出译文，不要有任何额外说明。
 ```
 
 With `--source auto`, `{source-language}` is omitted. Prompt length, source text, and generated output all count toward the served context. Parameters left unspecified above are not assigned new fixed values by this model card.
@@ -216,15 +222,17 @@ text = response.choices[0].message.content or ""
 if "</think>" in text: text = text.split("</think>", 1)[1]
 print(text.strip().removeprefix("<think>").strip())
 
-# 2. Translation with terminology (hard constraint) and style instruction (soft constraint)
-prompt_with_constraints = """请将以下文本翻译为英语。
-翻译要求：保持专业严谨的技术文档风格
-请严格遵循以下术语对应关系进行翻译：
-- 大模型 -> LLM
-- 开源 -> open-source
-直接输出翻译结果，不要进行任何解释。
+# 2. instTrans constrained translation (hard glossary + soft style constraints)
+prompt_with_constraints = """请将以下专栏文章翻译成英语，并且严格遵循所有约束要求。
+
+【源文】
+王平仲采用了更加昂贵的碳纤维材料。碳纤维的好处就是它抗裂缝，而且能显著提升混凝土的刚性。
+
+【约束要求】
+1. 【硬性要求】专名/术语对照: 碳纤维→carbon fiber、抗裂缝→crack resistance、刚性→stiffness
+2. 【注意】保持专业严谨的技术工程文档风格
 
-这是一个大模型开源项目。"""
+只输出译文，不要有任何额外说明。"""
 
 response = client.chat.completions.create(
     model="IndexTeam/Index-Translate-2B",
@@ -252,17 +260,22 @@ python inference/llm/translate.py \
   "你好，世界。今天天气不错。" --target en \
   --model IndexTeam/Index-Translate-2B
 
-# Translation with terminology (glossary) and style instruction
+# Hard constraint: Terminology glossary (-g / --glossary)
+python inference/llm/translate.py \
+  "王平仲采用了更加昂贵的碳纤维材料。碳纤维的好处就是它抗裂缝。" --target en \
+  -g "碳纤维:carbon fiber, 抗裂缝:crack resistance" \
+  --model IndexTeam/Index-Translate-2B
+
+# Hard constraint: Structured data / JSON format preservation (-H / --hard)
 python inference/llm/translate.py \
-  "这是一个大模型开源项目，欢迎开发者体验。" --target en \
-  --glossary "大模型:LLM" "开源:open-source" \
-  --instruction "保持专业严谨的技术文档风格" \
+  '{"user_id": 1024, "message": "您的订��已支付完成。"}' --target en \
+  -H "保留源文中的 JSON 格式标记不变" \
   --model IndexTeam/Index-Translate-2B
 
-# Structured data translation (JSON preservation)
+# Soft constraint: Style (-S / --soft) & Genre (-d / --genre)
 python inference/llm/translate.py \
-  '{"title": "公告", "content": "系统升级中"}' --target en \
-  --instruction "请保持原有JSON结构不变，只翻译字段值" \
+  "今天下午的会议临时取消了，改天我们再碰一下商量。" --target en \
+  -d "商务邮件" -S "调整为严谨、正式、礼貌的商务公文风格" \
   --model IndexTeam/Index-Translate-2B
 ```