Scalable Frequency- and Length-Aware Subdocument Deduplication for Large Language Model Pretraining
Hai Wang, Chenhao Wang, Qifeng Cai, Yixiu Liu, Miao Peng
実装難易度
Hard
推論・学習コスト
High
想定用途
検出
概要
Large-scale pretraining corpora contain substantial duplicate content. Although document-level deduplication is widely used, removing subdocument-level redundancy remains challenging. At corpus scale, suffix-array-based methods are commonly applied independently within shards, leaving cross-shard duplicates undetected and making the resulting retention behavior sensitive to the sharding…
何が新しいか
Large-scale pretraining corpora contain substantial duplicate content. Although document-level deduplication is widely used, removing subdocument-level redundancy remains challenging.
何に使えるか
検出
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
A Unified 2D Framework for DeepLesion Detection, Segmentation and Short Report Generation
In previous work, we integrated large language models (LLMs) into the lesion segmentation model based on the U
SpatialCLI: Learning to Reason With Spatial Tools, Then Without Them
Vision-language models (VLMs) are increasingly used in embodied agents to interpret visual inputs, reason abou
MAFIA: Query-Only Memory Attacks via Probing and Factual Injection against Audited LLM Agents
LLMアグエントのメモリモジュールを悪用する記憶毒素攻撃を検討し、現実的な環境でのこの種の攻撃の効果を評価します。
LACE: Large Language Model Aided Multi-Agent Framework for Agile RISC-V Instruction Extension
Domain-specific Instruction Set Architecture eXtensions (ISAX) are widely adopted in the RISC-V ecosystem to a