Paved with True Intents: Intent-Aware Training Improves LLM Safety Classification Across Training Regimes
Jeremias Ferrao, Niclas Müller-Hof, Iustin Sîrbu, Traian Rebedea, Yftah Ziser
実装難易度
Hard
推論・学習コスト
High
想定用途
分類
概要
We argue that safety classifiers should model user intent as an explicit signal between the prompt and the final label. To study this, we introduce AIMS, a human-annotated dataset of 1,724 difficult safety prompts, each paired with an intent description and harm label. We use AIMS to evaluate intent-aware training across supervised fine-tuning, preference learning, reasoning distillation, and…
何が新しいか
We argue that safety classifiers should model user intent as an explicit signal between the prompt and the final label. To study this, we introduce AIMS, a human-annotated dataset of 1,724 difficult…
何に使えるか
分類
実装情報
- Paper URL
- あり
実装チェックリスト
実装または配布ページ
要確認Paper onlyの可能性があるため再実装前提で確認してください。
一次情報リンク
OKPaper
検証しやすさ
要確認公式実装が見つからないため、論文から再実装する前提です。
計算資源
要確認学習や高解像度推論ではGPUメモリと実行時間に注意が必要です。
ライセンス
未取得配布元のLICENSE、モデルカード、Paperの利用条件を確認してください。
商用利用
未取得研究利用限定、データセット由来制限、API規約の有無を確認してください。
自社データで試すなら
製造業・材料開発のExcel/CSVデータに落とし込むための最初の手順です。
- 1まず自社データを、入力条件、目的変数、評価したい指標に分けて整理します。
- 2LightGBMやRandom Forestなどのベースラインを先に作り、この手法と比較します。
- 3評価指標はR2/RMSE、AUC、異常検知の再現率、実験回数削減率など、現場の意思決定に近いものを選びます。
- 4SHAPや特徴量重要度で、効いている因子が物理・化学・工程知識と矛盾しないか確認します。
実装難易度
Hard - 公式実装が見つからないため、論文から再実装する前提です。
必要リソース
- GPU目安: High
- データセット: 論文・リポジトリ側の指定を確認してください。
- 学習要否: 再学習や評価環境の準備が必要になる可能性があります。
- 学習や高解像度推論ではGPUメモリと実行時間に注意が必要です。
実務で使う場合の注意点
- ライセンスと商用利用条件は、Paper / GitHub / Hugging Face の配布元で確認してください。
- 精度、再現性、計算コストはデータセットや評価条件に依存します。
- 個人情報や機密データを扱う場合は、入力データの保存先と外部API利用条件を確認してください。
関連記事
MLFFM-SegDiff: A Multi-Level Feature Fusion Diffusion Model for Skin Lesion Segmentation
Skin lesion segmentation is a key task in computer-aided dermatological diagnosis, where accuracy directly imp
TaskTok: Delving into Task Tokens for Task-driven Image Restoration
While traditional image restoration focuses on perceptual quality, Task-Driven Image Restoration (TDIR) aims t
Dual Distribution Estimation for Zero-shot Noisy Test-Time Adaptation with VLMs
類似性のない分布の異常値は、実務上は問題を引き起こしますが、現行のテスト時適応 (TTA) は、学習時の分布と異なるデータに高い弱点を持つ限り、未知のデータへの適用は問題です。現行のゼロショット TT における、弱い分布
TACO: Towards Task-Consistent Open-Vocabulary Adaptation in Video Recognition
CLIP に基づいてオープン言語化が適合しない問題があります。