TDHook: A Lightweight Framework for Interpretability

• URL: http://arxiv.org/abs/2509.25475v1
• Date: Mon, 29 Sep 2025 20:28:43 GMT
• Title: TDHook: A Lightweight Framework for Interpretability
• Authors: Yoann Poupart,
• Abstract summary: Interpretability of Deep Neural Networks (DNNs) is a growing field driven by the study of vision and language models.<n>We present TDHook, an open-source, lightweight, generic interpretability framework based on $texttttensordict$.<n>It focuses on handling complex composed models which can be trained for Computer Vision, Natural Language Processing, Reinforcement Learning or any other domain.
• Score: 0.0
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Interpretability of Deep Neural Networks (DNNs) is a growing field driven by the study of vision and language models. Yet, some use cases, like image captioning, or domains like Deep Reinforcement Learning (DRL), require complex modelling, with multiple inputs and outputs or use composable and separated networks. As a consequence, they rarely fit natively into the API of popular interpretability frameworks. We thus present TDHook, an open-source, lightweight, generic interpretability framework based on $\texttt{tensordict}$ and applicable to any $\texttt{torch}$ model. It focuses on handling complex composed models which can be trained for Computer Vision, Natural Language Processing, Reinforcement Learning or any other domain. This library features ready-to-use methods for attribution, probing and a flexible get-set API for interventions, and is aiming to bridge the gap between these method classes to make modern interpretability pipelines more accessible. TDHook is designed with minimal dependencies, requiring roughly half as much disk space as $\texttt{transformer_lens}$, and, in our controlled benchmark, achieves up to a $\times$2 speed-up over $\texttt{captum}$ when running integrated gradients for multi-target pipelines on both CPU and GPU. In addition, to value our work, we showcase concrete use cases of our library with composed interpretability pipelines in Computer Vision (CV) and Natural Language Processing (NLP), as well as with complex models in DRL.

Related papers

• $\ exttt{lrnnx}$: A library for Linear RNNs [0.0]
  Linear recurrent neural networks (LRNNs) provide a structured approach to sequence modeling that bridges classical linear dynamical systems and modern deep learning, offering both expressive power and theoretical guarantees on stability and trainability.<n>Existing implementations are fragmented across different software frameworks, often rely on framework-specific optimizations, and in some cases require custom kernels or lack publicly available code altogether.<n>We introduce $textttlrn$, a unified software library that implements several modern LRNN architectures under a common interface.
  arXiv  Detail & Related papers  (2026-02-09T15:48:48Z)
• GLiNER2: An Efficient Multi-Task Information Extraction System with Schema-Driven Interface [0.873811641236639]
  We present GLiNER2, a unified framework that enhances the original GLiNER architecture to support named entity recognition, text classification, and hierarchical structured data extraction.<n>Our experiments demonstrate competitive performance across extraction and classification tasks with substantial improvements in deployment accessibility.
  arXiv  Detail & Related papers  (2025-07-24T16:11:14Z)
• NNTile: a machine learning framework capable of training extremely large GPT language models on a single node [83.9328245724548]
  NNTile is based on a StarPU library, which implements task-based parallelism and schedules all provided tasks onto all available processing units.<n>It means that a particular operation, necessary to train a large neural network, can be performed on any of the CPU cores or GPU devices.
  arXiv  Detail & Related papers  (2025-04-17T16:22:32Z)
• Position-Aware Depth Decay Decoding ($D^3$): Boosting Large Language Model Inference Efficiency [26.173523821684306]
  A token-position aware layer skipping framework is proposed to save 1.5x times operations efficiently while maintaining performance.<n>Experiments on large language models with $7 sim 70$ billion parameters show that $D3$ can achieve an average 1.5x speedup compared with the full-inference pipeline.
  arXiv  Detail & Related papers  (2025-03-11T15:15:54Z)
• NNsight and NDIF: Democratizing Access to Open-Weight Foundation Model Internals [58.83169560132308]
  We introduce NNsight and NDIF, technologies that work in tandem to enable scientific study of the representations and computations learned by very large neural networks.
  arXiv  Detail & Related papers  (2024-07-18T17:59:01Z)
• pyvene: A Library for Understanding and Improving PyTorch Models via Interventions [79.72930339711478]
  $textbfpyvene$ is an open-source library that supports customizable interventions on a range of different PyTorch modules. We show how $textbfpyvene$ provides a unified framework for performing interventions on neural models and sharing the intervened upon models with others.
  arXiv  Detail & Related papers  (2024-03-12T16:46:54Z)
• Speculative Decoding with Big Little Decoder [108.95187338417541]
  Big Little Decoder (BiLD) is a framework that can improve inference efficiency and latency for a wide range of text generation applications. On an NVIDIA T4 GPU, our framework achieves a speedup of up to 2.12x speedup with minimal generation quality degradation. Our framework is fully plug-and-play and can be applied without any modifications in the training process or model architecture.
  arXiv  Detail & Related papers  (2023-02-15T18:55:29Z)
• Scaling Up Models and Data with $\texttt{t5x}$ and $\texttt{seqio}$ [118.04625413322827]
  $texttt5x$ and $texttseqio$ are open source software libraries for building and training language models. These libraries have been used to train models with hundreds of billions of parameters on datasets with multiple terabytes of training data.
  arXiv  Detail & Related papers  (2022-03-31T17:12:13Z)
• Improving Robustness and Generality of NLP Models Using Disentangled Representations [62.08794500431367]
  Supervised neural networks first map an input $x$ to a single representation $z$, and then map $z$ to the output label $y$. We present methods to improve robustness and generality of NLP models from the standpoint of disentangled representation learning. We show that models trained with the proposed criteria provide better robustness and domain adaptation ability in a wide range of supervised learning tasks.
  arXiv  Detail & Related papers  (2020-09-21T02:48:46Z)
• CLEVR Parser: A Graph Parser Library for Geometric Learning on Language Grounded Image Scenes [2.750124853532831]
  CLEVR dataset has been used extensively in language grounded visual reasoning in Machine Learning (ML) and Natural Language Processing (NLP) domains. We present a graph library for CLEVR that provides functionalities for object-centric attributes and relationships extraction, and construction of structural graph representations for dual modalities. We discuss downstream usage and applications of the library, and how it accelerates research for the NLP research community.
  arXiv  Detail & Related papers  (2020-09-19T03:32:37Z)
• Captum: A unified and generic model interpretability library for PyTorch [49.72749684393332]
  We introduce a novel, unified, open-source model interpretability library for PyTorch. The library contains generic implementations of a number of gradient and perturbation-based attribution algorithms. It can be used for both classification and non-classification models.
  arXiv  Detail & Related papers  (2020-09-16T18:57:57Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.