---
language: [en]
license: mit
tags: [conceptual-metaphor, annotation, lakoff, parallel-corpus, lora, qwen2.5]
base_model: Qwen/Qwen2.5-1.5B-Instruct
---

# CM-Annotator (English) — Conceptual Metaphor Annotator

LoRA adapter that annotates English sentences with conceptual metaphor spans
(Lakoff & Johnson tradition) over a curated closed-set inventory (215 entries, based on the
Master Metaphor List, Lakoff et al. 1991). Outputs strict JSON:
`{"has_metaphor": bool, "metaphor_spans": [{text, cm_id, cm_name, source_domain, target_domain,
conventionality, confidence}]}`. Sentences without conceptual metaphors yield empty spans
(Null-CM negatives are ~30% of training).

## Training (distillation)

- Silver labels distilled from a local LM Studio teacher (qwen3.6-35b-a3b), k=2 self-consistency;
  only agreeing labels kept. Details: `docs/TRAINING.md`. Silver labels are NOT gold-human labels.
- Base: `Qwen/Qwen2.5-1.5B-Instruct` · LoRA r=16 α=32 ·
  completion-only loss · seed=42 · trained 2026-11-07T18:00:10.229852+00:00

## Evaluation (dev vs silver, greedy)

| metric | value |
|---|---|
| n_eval | 120 |
| JSON valid rate | 1.0 |
| span precision / recall / F1 | 0.0962 / 0.1111 / 0.1031 |

## Usage

Load base + adapter with peft; build prompts exactly as in `service/inference.py`
(system rules + INVENTORY block + SENTENCE). Trained-on data derives from UM-Corpus (CC BY-NC-ND):
training data is not redistributed; reproduce via the pipeline in the repository.

## Limitations

Precision bounded by teacher quality (see audit table in docs/TRAINING.md); recall conservative
on conventional metaphors; confidence not calibrated; Chinese side corpus is mostly simplified script.