Goto

Collaborating Authors

 tbm


Interpretable Discriminative Text Representations via Agreement and Label Disentanglement

arXiv.org Machine Learning

Interpretable text representations should expose coordinates that are not only predictive, but also meaningful enough for independent auditors to apply. Existing discriminative representations often use anonymous embedding directions, while concept-bottleneck and LLM-assisted methods attach natural-language names to features without ensuring that those definitions are reproducible or distinct from the target label. We propose an operational criterion for interpretable discriminative text representations: each coordinate should satisfy conceptual clarity, measured by chance-adjusted agreement between independent annotators applying the feature definition, and label disentanglement, meaning the feature should not merely paraphrase the prediction target. We instantiate this criterion in LLM-assisted Feature Discovery (LFD), an iterative method that proposes lexical and semantic features from contrastive outcome-opposed text pairs, screens candidates using cross-LLM Cohen's $ฮบ$, and selects features by residual held-out predictive gain. A stylized analysis connects the $ฮบ$ screen to a per-feature annotation-noise bound, formalizing agreement as a reliability check. Across ten text-classification tasks spanning seven corpora, LFD matches the predictive performance of a strong text bottleneck baseline while producing substantially clearer and less label-entangled features. Human audits with 232 raters show that LFD features achieve higher human--human and human--LLM agreement than baseline concepts, and raters consistently judge them as less label-leaking. These results suggest that agreement-tested, label-disentangled coordinates provide a practical auditability standard for interpretable text classification.




To Reviewer 1

Neural Information Processing Systems

We thank the reviewers for the helpful comments and feedback. Our responses are detailed below. We will make the suggested edits for clarity. The improved interpretability with little loss of accuracy makes the sparse TBM appealing in applications. We agree with reviewer that MSE is not the best metric for clustering.


The Trajectory Bundle Method: Unifying Sequential-Convex Programming and Sampling-Based Trajectory Optimization

arXiv.org Artificial Intelligence

We present a unified framework for solving trajectory optimization problems in a derivative-free manner through the use of sequential convex programming. Traditionally, nonconvex optimization problems are solved by forming and solving a sequence of convex optimization problems, where the cost and constraint functions are approximated locally through Taylor series expansions. This presents a challenge for functions where differentiation is expensive or unavailable. In this work, we present a derivative-free approach to form these convex approximations by computing samples of the dynamics, cost, and constraint functions and letting the solver interpolate between them. Our framework includes sample-based trajectory optimization techniques like model-predictive path integral (MPPI) control as a special case and generalizes them to enable features like multiple shooting and general equality and inequality constraints that are traditionally associated with derivative-based sequential convex programming methods. The resulting framework is simple, flexible, and capable of solving a wide variety of practical motion planning and control problems.


Tree-based Models for Vertical Federated Learning: A Survey

arXiv.org Artificial Intelligence

Tree-based models have achieved great success in a wide range of real-world applications due to their effectiveness, robustness, and interpretability, which inspired people to apply them in vertical federated learning (VFL) scenarios in recent years. In this paper, we conduct a comprehensive study to give an overall picture of applying tree-based models in VFL, from the perspective of their communication and computation protocols. We categorize tree-based models in VFL into two types, i.e., feature-gathering models and label-scattering models, and provide a detailed discussion regarding their characteristics, advantages, privacy protection mechanisms, and applications. This study also focuses on the implementation of tree-based models in VFL, summarizing several design principles for better satisfying various requirements from both academic research and industrial deployment. We conduct a series of experiments to provide empirical observations on the differences and advances of different types of tree-based models.


Consistent spectral clustering in sparse tensor block models

arXiv.org Artificial Intelligence

High-order clustering aims to classify objects in multiway datasets that are prevalent in various fields such as bioinformatics, social network analysis, and recommendation systems. These tasks often involve data that is sparse and high-dimensional, presenting significant statistical and computational challenges. This paper introduces a tensor block model specifically designed for sparse integer-valued data tensors. We propose a simple spectral clustering algorithm augmented with a trimming step to mitigate noise fluctuations, and identify a density threshold that ensures the algorithm's consistency. Our approach models sparsity using a sub-Poisson noise concentration framework, accommodating heavier than sub-Gaussian tails. Remarkably, this natural class of tensor block models is closed under aggregation across arbitrary modes. Consequently, we obtain a comprehensive framework for evaluating the tradeoff between signal loss and noise reduction during data aggregation. The analysis is based on a novel concentration bound for sparse random Gram matrices. The theoretical findings are illustrated through simulation experiments.


Interpretable-by-Design Text Classification with Iteratively Generated Concept Bottleneck

arXiv.org Artificial Intelligence

Deep neural networks excel in text classification tasks, yet their application in high-stakes domains is hindered by their lack of interpretability. To address this, we propose Text Bottleneck Models (TBMs), an intrinsically interpretable text classification framework that offers both global and local explanations. Rather than directly predicting the output label, TBMs predict categorical values for a sparse set of salient concepts and use a linear layer over those concept values to produce the final prediction. These concepts can be automatically discovered and measured by a Large Language Model (LLM), without the need for human curation. On 12 diverse datasets, using GPT-4 for both concept generation and measurement, we show that TBMs can rival the performance of established black-box baselines such as GPT-4 fewshot and finetuned DeBERTa, while falling short against finetuned GPT-3.5. Overall, our findings suggest that TBMs are a promising new framework that enhances interpretability, with minimal performance tradeoffs, particularly for general-domain text.


Swarms of Mini Robots Could Dig the Tunnels of the Future

WIRED

For decades, engineers seeking to build tunnels underground have relied on huge tube-like machines armed with a frightening array of cutting wheels at one end--blades that eat dirt for breakfast. These behemoths, called tunnel-boring machines, or TBMs, are expensive and often custom-built for each project, as were the TBMs used to excavate a path for London's recently opened Elizabeth Line railway. The machines deployed on that project weighed over 1,000 tons each and cut tunnels over 7 meters in diameter beneath the UK capital. But British startup hyperTunnel has other ideas. The firm proposes a future in which much smaller, roughly 3-meter-long robots shaped like half-cylinders zoom about underground via predrilled pipes.


Intelligent decision-making method of TBM operating parameters based on multiple constraints and objective optimization

arXiv.org Artificial Intelligence

The decision-making of TBM operating parameters has an important guiding significance for TBM safe and efficient construction, and it has been one of the research hotpots in the field of TBM tunneling. For this purpose, this paper introduces rock-breaking rules into machine learning method, and a rock-machine mapping dual-driven by physical-rule and data-mining is established with high accuracy. This dual-driven mappings are subsequently used as objective function and constraints to build a decision-making method for TBM operating parameters. By searching the revolution per minute and penetration corresponding to the extremum of the objective function subject to the constraints, the optimal operating parameters can be obtained. This method is verified in the field of the Second Water Source Channel of Hangzhou, China, resulting in the average penetration rate increased by 11.3%, and the total cost decreased by 10.0%, which proves the practicability and effectiveness of the developed decision-making model.