# Token-wise Influential Training Data Retrieval for Large Language Models

arXiv: https://arxiv.org/abs/2405.11724 · HTML: https://mlsanity.com/paper/2405.11724/

## Key facts

- Cited by 20 papers in the map (data through 14 Oct 2026).
- Builds on 31 of its 68 references that are in the map.
- First posted on arXiv 20 May 2024.

## Builds on (most cited first, 30 of 31)

- [Llama 2: Open Foundation and Fine-Tuned Chat Models](https://mlsanity.com/paper/2307.09288/) (2023)
- [QLoRA: Efficient Finetuning of Quantized LLMs](https://mlsanity.com/paper/2305.14314/) (2023)
- [Self-Instruct: Aligning Language Models with Self-Generated Instructions](https://mlsanity.com/paper/2212.10560/) (2022)
- [Understanding Black-box Predictions via Influence Functions](https://mlsanity.com/paper/1703.04730/) (2017)
- [Estimating Training Data Influence by Tracing Gradient Descent](https://mlsanity.com/paper/2002.08484/) (2020)
- [Scalable Extraction of Training Data from (Production) Language Models](https://mlsanity.com/paper/2311.17035/) (2023)
- [Multi-News: a Large-Scale Multi-Document Summarization Dataset and Abstractive Hierarchical Model](https://mlsanity.com/paper/1906.01749/) (2019)
- [Describe, Explain, Plan and Select: Interactive Planning with Large Language Models Enables Open-World Multi-Task Agents](https://mlsanity.com/paper/2302.01560/) (2023)
- [JudgeLM: Fine-tuned Large Language Models are Scalable Judges](https://mlsanity.com/paper/2310.17631/) (2023)
- [Studying Large Language Model Generalization with Influence Functions](https://mlsanity.com/paper/2308.03296/) (2023)
- [Deduplicating Training Data Mitigates Privacy Risks in Language Models](https://mlsanity.com/paper/2202.06539/) (2022)
- [Influence Functions in Deep Learning Are Fragile](https://mlsanity.com/paper/2006.14651/) (2020)
- [What Does it Mean for a Language Model to Preserve Privacy?](https://mlsanity.com/paper/2202.05520/) (2022)
- [Representer Point Selection for Explaining Deep Neural Networks](https://mlsanity.com/paper/1811.09720/) (2018)
- [CulturaX: A Cleaned, Enormous, and Multilingual Dataset for Large Language Models in 167 Languages](https://mlsanity.com/paper/2309.09400/) (2023)
- [Challenges in Detoxifying Language Models](https://mlsanity.com/paper/2109.07445/) (2021)
- [Scaling Laws and Interpretability of Learning from Repeated Data](https://mlsanity.com/paper/2205.10487/) (2022)
- [Explaining Black Box Predictions and Unveiling Data Artifacts through Influence Functions](https://mlsanity.com/paper/2005.06676/) (2020)
- [If Influence Functions are the Answer, Then What is the Question?](https://mlsanity.com/paper/2209.05364/) (2022)
- [FastIF: Scalable Influence Functions for Efficient Model Interpretation and Debugging](https://mlsanity.com/paper/2012.15781/) (2020)
- [Backdoor Attacks for In-Context Learning with Language Models](https://mlsanity.com/paper/2307.14692/) (2023)
- [Prompt as Triggers for Backdoor Attack: Examining the Vulnerability in Language Models](https://mlsanity.com/paper/2305.01219/) (2023)
- [AlpaCare:Instruction-tuned Large Language Models for Medical Application](https://mlsanity.com/paper/2310.14558/) (2023)
- [Data Cleansing for Models Trained with SGD](https://mlsanity.com/paper/1906.08473/) (2019)
- [Simfluence: Modeling the Influence of Individual Training Examples by Simulating Training Runs](https://mlsanity.com/paper/2303.08114/) (2023)
- [Text Detoxification using Large Pre-trained Neural Models](https://mlsanity.com/paper/2109.08914/) (2021)
- [Steering Large Language Models for Machine Translation with Finetuning and In-Context Learning](https://mlsanity.com/paper/2310.13448/) (2023)
- [Fast and Accurate Network Embeddings via Very Sparse Random Projection](https://mlsanity.com/paper/1908.11512/) (2019)
- [Billion-scale Network Embedding with Iterative Random Projection](https://mlsanity.com/paper/1805.02396/) (2018)
- [Contrastive Error Attribution for Finetuned Language Models](https://mlsanity.com/paper/2212.10722/) (2022)

## Built on by (most cited first, 20 of 20)

- [Do Influence Functions Work on Large Language Models?](https://mlsanity.com/paper/2409.19998/) (2024)
- [UniGuardian: A Unified Defense for Detecting Prompt Injection, Backdoor Attacks and Adversarial Attacks in Large Language Models](https://mlsanity.com/paper/2502.13141/) (2025)
- [A Snapshot of Influence: A Local Data Attribution Framework for Online Reinforcement Learning](https://mlsanity.com/paper/2505.19281/) (2025)
- [GraSS: Scalable Data Attribution with Gradient Sparsification and Sparse Projection](https://mlsanity.com/paper/2505.18976/) (2025)
- [All You Need is One: Capsule Prompt Tuning with a Single Vector](https://mlsanity.com/paper/2510.16670/) (2025)
- [VTBench: Evaluating Visual Tokenizers for Autoregressive Image Generation](https://mlsanity.com/paper/2505.13439/) (2025)
- [Where Did It Go Wrong? Attributing Undesirable LLM Behaviors via Representation Gradient Tracing](https://mlsanity.com/paper/2510.02334/) (2025)
- [DMin: Scalable Training Data Influence Estimation for Diffusion Models](https://mlsanity.com/paper/2412.08637/) (2024)
- [BLISS: A Lightweight Bilevel Influence Scoring Method for Data Selection in Language Model Pretraining](https://mlsanity.com/paper/2510.06048/) (2025)
- [TRIM: Token-wise Attention-Derived Saliency for Data-Efficient Instruction Tuning](https://mlsanity.com/paper/2510.07118/) (2025)
- [Cite Pretrain: Retrieval-Free Knowledge Attribution for Large Language Models](https://mlsanity.com/paper/2506.17585/) (2025)
- [Randomized Antipodal Search Done Right for Data Pareto Improvement of LLM Unlearning](https://mlsanity.com/paper/2604.16591/) (2026)
- [Online Gradient Boosting Decision Tree: In-Place Updates for Efficient Adding/Deleting Data](https://mlsanity.com/paper/2502.01634/) (2025)
- [On Representation Redundancy in Large-Scale Instruction Tuning Data Selection](https://mlsanity.com/paper/2602.13773/) (2026)
- [Sketching the Readout of Large Language Models for Scalable Data Attribution and Valuation](https://mlsanity.com/paper/2604.16197/) (2026)
- [Estimating the Effects of Sample Training Orders for Large Language Models without Retraining](https://mlsanity.com/paper/2505.22042/) (2025)
- [Data Security in Large Language Models: Risks, Defense, and Directions](https://mlsanity.com/paper/2508.02312/) (2025)
- [RapidUn: Influence-Driven Parameter Reweighting for Efficient Large Language Model Unlearning](https://mlsanity.com/paper/2512.04457/) (2025)
- [Select or Project? Evaluating Lower-dimensional Vectors for LLM Training Data Explanations](https://mlsanity.com/paper/2601.16651/) (2026)
- [Which Influence Are We Estimating? The Role of Counterfactual Specifications in Data Attribution](https://mlsanity.com/paper/2609.31214/) (2026)

## Abstract (arXiv)

Given a Large Language Model (LLM) generation, how can we identify which training data led to this generation? In this paper, we proposed RapidIn, a scalable framework adapting to LLMs for estimating the influence of each training data. The proposed framework consists of two stages: caching and retrieval. First, we compress the gradient vectors by over 200,000x, allowing them to be cached on disk or in GPU/CPU memory. Then, given a generation, RapidIn efficiently traverses the cached gradients to estimate the influence within minutes, achieving over a 6,326x speedup. Moreover, RapidIn supports multi-GPU parallelization to substantially accelerate caching and retrieval. Our empirical result confirms the efficiency and effectiveness of RapidIn.