# The Attribution Contract for Generative Language Models

arXiv: https://arxiv.org/abs/2605.23080 · HTML: https://mlsanity.com/paper/2605.23080/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Placed in MoE large language models, one of 12 directions in Mixture-of-experts.
- Builds on 33 of its 36 references that are in the map.
- First posted on arXiv 21 May 2026.

## Placed in

- Mixture-of-experts / [MoE large language models](https://mlsanity.com/moe/moe-large-language-models/)

## Builds on (most cited first, 30 of 33)

- [A Unified Approach to Interpreting Model Predictions](https://mlsanity.com/paper/1705.07874/) (2017)
- [Axiomatic Attribution for Deep Networks](https://mlsanity.com/paper/1703.01365/) (2017)
- [Efficient Streaming Language Models with Attention Sinks](https://mlsanity.com/paper/2309.17453/) (2023)
- [Survey of Hallucination in Natural Language Generation](https://mlsanity.com/paper/2202.03629/) (2022)
- [Mixtral of Experts](https://mlsanity.com/paper/2401.04088/) (2024)
- [A Survey on Hallucination in Large Language Models: Principles, Taxonomy, Challenges, and Open Questions](https://mlsanity.com/paper/2311.05232/) (2023)
- [Rethinking the Role of Demonstrations: What Makes In-Context Learning Work?](https://mlsanity.com/paper/2202.12837/) (2022)
- [Red Teaming Language Models with Language Models](https://mlsanity.com/paper/2202.03286/) (2022)
- [Large Language Diffusion Models](https://mlsanity.com/paper/2502.09992/) (2025)
- [Discrete Diffusion Modeling by Estimating the Ratios of the Data Distribution](https://mlsanity.com/paper/2310.16834/) (2023)
- [Red Teaming Language Models to Reduce Harms: Methods, Scaling Behaviors, and Lessons Learned](https://mlsanity.com/paper/2209.07858/) (2022)
- [The Mythos of Model Interpretability](https://mlsanity.com/paper/1606.03490/) (2016)
- [Towards Faithfully Interpretable NLP Systems: How should we define and evaluate faithfulness?](https://mlsanity.com/paper/2004.03685/) (2020)
- [Evaluating the visualization of what a Deep Neural Network has learned](https://mlsanity.com/paper/1509.06321/) (2015)
- [OLMoE: Open Mixture-of-Experts Language Models](https://mlsanity.com/paper/2409.02060/) (2024)
- [True to the Model or True to the Data?](https://mlsanity.com/paper/2006.16234/) (2020)
- [ContextCite: Attributing Model Generation to Context](https://mlsanity.com/paper/2409.00729/) (2024)
- [Interpreting Language Models with Contrastive Explanations](https://mlsanity.com/paper/2202.10419/) (2022)
- [The effectiveness of feature attribution methods and its correlation with automatic evaluation scores](https://mlsanity.com/paper/2105.14944/) (2021)
- [Impossibility Theorems for Feature Attribution](https://mlsanity.com/paper/2212.11870/) (2022)
- [Inseq: An Interpretability Toolkit for Sequence Generation Models](https://mlsanity.com/paper/2302.13942/) (2023)
- [Post hoc Explanations may be Ineffective for Detecting Unknown Spurious Correlation](https://mlsanity.com/paper/2212.04629/) (2022)
- [Sequential Integrated Gradients: a simple but effective method for explaining language models](https://mlsanity.com/paper/2305.15853/) (2023)
- [Discretized Integrated Gradients for Explaining Language Models](https://mlsanity.com/paper/2108.13654/) (2021)
- [Using Captum to Explain Generative Language Models](https://mlsanity.com/paper/2312.05491/) (2023)
- [Continuous Diffusion Model for Language Modeling](https://mlsanity.com/paper/2502.11564/) (2025)
- [Interpretable LLM-based Table Question Answering](https://mlsanity.com/paper/2412.12386/) (2024)
- [ReAGent: A Model-agnostic Feature Attribution Method for Generative Language Models](https://mlsanity.com/paper/2402.00794/) (2024)
- [The Dead Salmons of AI Interpretability](https://mlsanity.com/paper/2512.18792/) (2025)
- [Faithfulness Measurable Masked Language Models](https://mlsanity.com/paper/2310.07819/) (2023)

## Abstract (arXiv)

Feature attribution scores each part of an input by how much it explains a model's output. We argue that in generative language models these scores carry no fixed meaning. A classifier has a single output to explain, but a generative model produces its output token by token, and each generated token is both an output and an input, so explaining the output becomes several distinct questions. We support this claim by proposing the Attribution Contract, a framework that names the question a set of attribution scores answers. A contract specifies the model score being explained, what is held fixed, the target output, the generation process, and which features can receive attribution, choices that matter in practice. For example, eligible features decide what a score explains: under a local next-token contract, the model's own generated tokens receive 29% of the attribution mass, comparable to the 25% on the input question, and reading this as a prompt-level explanation is a mistake we name the self-attribution fallacy. The generative process also shapes feature attribution: on a mixture-of-experts model, Integrated Gradients completeness is unreliable, since recomputing expert routing along the attribution path leaves a large residual that holding the routing fixed substantially lowers, while a masked-diffusion model, which fixes its choices at generation, has a residual that instead decreases with more steps. These results show that Attribution Contracts should be carefully specified when proposing and evaluating attribution methods for generative language models, since the meaning and evaluation of an attribution depend on its explanatory setting.