# Beyond Shapley: An Influence-Based Data Auditing Pipeline for LLM Alignment and Evaluation

arXiv: https://arxiv.org/abs/2607.22766 · HTML: https://mlsanity.com/paper/2607.22766/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Placed in LLM preference optimization methods, one of 10 directions in LLM alignment / RLHF / preference optimisation.
- Builds on 19 of its 26 references that are in the map.
- First posted on arXiv 24 Jul 2026.

## Placed in

- LLM alignment / RLHF / preference optimisation / [LLM preference optimization methods](https://mlsanity.com/alignment/llm-preference-optimization-methods/)

## Builds on (most cited first, 19 of 19)

- [Training language models to follow instructions with human feedback](https://mlsanity.com/paper/2203.02155/) (2022)
- [Judging LLM-as-a-Judge with MT-Bench and Chatbot Arena](https://mlsanity.com/paper/2306.05685/) (2023)
- [Direct Preference Optimization: Your Language Model is Secretly a Reward Model](https://mlsanity.com/paper/2305.18290/) (2023)
- [Training a Helpful and Harmless Assistant with Reinforcement Learning from Human Feedback](https://mlsanity.com/paper/2204.05862/) (2022)
- [Gemma 3 Technical Report](https://mlsanity.com/paper/2503.19786/) (2025)
- [Qwen3 Embedding: Advancing Text Embedding and Reranking Through Foundation Models](https://mlsanity.com/paper/2506.05176/) (2025)
- [Large Language Models are not Fair Evaluators](https://mlsanity.com/paper/2305.17926/) (2023)
- [M3-Embedding: Multi-Linguality, Multi-Functionality, Multi-Granularity Text Embeddings Through Self-Knowledge Distillation](https://mlsanity.com/paper/2402.03216/) (2024)
- [Efficient and robust approximate nearest neighbor search using Hierarchical Navigable Small World graphs](https://mlsanity.com/paper/1603.09320/) (2016)
- [Deduplicating Training Data Makes Language Models Better](https://mlsanity.com/paper/2107.06499/) (2021)
- [Data Shapley: Equitable Valuation of Data for Machine Learning](https://mlsanity.com/paper/1904.02868/) (2019)
- [Textbooks Are All You Need](https://mlsanity.com/paper/2306.11644/) (2023)
- [SemDeDup: Data-efficient learning at web-scale through semantic deduplication](https://mlsanity.com/paper/2303.09540/) (2023)
- [HelpSteer2: Open-source dataset for training top-performing reward models](https://mlsanity.com/paper/2406.08673/) (2024)
- [Qwen3.5-Omni Technical Report](https://mlsanity.com/paper/2604.15804/) (2026)
- [Efficient Task-Specific Data Valuation for Nearest Neighbor Algorithms](https://mlsanity.com/paper/1908.08619/) (2019)
- [TokenShapley: Token Level Context Attribution with Shapley Value](https://mlsanity.com/paper/2507.05261/) (2025)
- [Fair Document Valuation in LLM Summaries via Shapley Values](https://mlsanity.com/paper/2505.23842/) (2025)
- [WaKA: Data Attribution using K-Nearest Neighbors and Membership Privacy Principles](https://mlsanity.com/paper/2411.01357/) (2024)

## Abstract (arXiv)

Post-training alignment of Large Language Models is increasingly bottlenecked by data quality: preference and instruction-tuning corpora accumulate label noise, safety inversions, and inter-annotator contradictions that deduplication misses and full-corpus LLM judging cannot afford. While Data Shapley offers a principled valuation framework, exact computation requires intractable retraining, and embedding surrogates like KNN-Shapley fail to separate high- from low-utility demonstrations (ρ = 0.0003 with generative utility). We introduce an inference-only valuation pipeline that computes the first-order term of an in-context cooperative game over semantic k-NN neighborhoods in O(k) forward passes without parameter updates, correlating positively with multi-shot Monte-Carlo in-context Shapley on 100% of targets at 19x less compute. Projecting pairwise log-likelihood shifts onto a directed influence graph characterizes each record via three signals: outgoing Exemplifying Advantage, incoming Exemplified Advantage, and a topological One-of-a-Kind Score protecting rare prompts. Pre-filtering restricts LLM twin arbitration to 1%-5% of records while achieving 4.5x enrichment in confirmed errors on HelpSteer2 and 2.2x on Anthropic HH-RLHF, catching contradictions that zero-shot likelihood and margin filters miss. During DPO training, confirmed contradictions oppose the update far more often than clean records (56.7% vs. 37.0% in LoRA gradients) and stay below chance in reward accuracy. Finally, auditing the HH-RLHF benchmark exposes label errors where aligned models disagree with human labels 73%-83% of the time, and retraining DPO after repairing or filtering flagged records improves reward accuracy by +0.5%-0.6% overall and +3.1%-3.5% on repaired inversions, even without LLM arbitration.