Pith. sign in

REVIEW 6 cited by

Editable Concept Bottleneck Models

Not yet reviewed by Pith; the record is open.

Re-run · record.json Download PDF Read on arXiv ↗

This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.

SPECIMEN: schema-true, not a live event

T0 review · schema-true

One-sentence machine reading of the paper's core claim.

pith:XXXXXXXX · record.json · timestamp

arxiv 2405.15476 v3 pith:N4VKXXBR submitted 2024-05-24 cs.LG cs.AIcs.CV

Lijie Hu , Chenyang Ren , Zhengyu Hu , Hongbin Lin , Cheng-Long Wang , Hui Xiong , Jingfeng Zhang , Di Wang This is my paper
classification cs.LGcs.AIcs.CV
keywords conceptcbmsdataecbmsbottleneckconceptseditablemodels
verification ladder T0 review T1 audit T2 compute T3 formal
0 comments
read the original abstract

Concept Bottleneck Models (CBMs) have garnered much attention for their ability to elucidate the prediction process through a humanunderstandable concept layer. However, most previous studies focused on cases where the data, including concepts, are clean. In many scenarios, we often need to remove/insert some training data or new concepts from trained CBMs for reasons such as privacy concerns, data mislabelling, spurious concepts, and concept annotation errors. Thus, deriving efficient editable CBMs without retraining from scratch remains a challenge, particularly in large-scale applications. To address these challenges, we propose Editable Concept Bottleneck Models (ECBMs). Specifically, ECBMs support three different levels of data removal: concept-label-level, concept-level, and data-level. ECBMs enjoy mathematically rigorous closed-form approximations derived from influence functions that obviate the need for retraining. Experimental results demonstrate the efficiency and adaptability of our ECBMs, affirming their practical value in CBMs.

Discussion (0). Sign in to comment.

Forward citations

Cited by 6 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score. Full citation record

  1. Efficient Text-Attributed Graph Learning through Selective Annotation and Graph Alignment

    cs.LG 2025-06 conditional novelty 6.0 of 10

    GAGA matches or exceeds state-of-the-art accuracy on several text-attributed graph benchmarks while requiring large language model annotations for only 1% of nodes or edges.

  2. Enhancing Interpretable Image Classification Through LLM Agents and Conditional Concept Bottleneck Models

    cs.CL 2025-06 conditional novelty 6.0 of 10

    Concept bottleneck models get a label-conditioned scoring layer and an LLM agent that iteratively refines the concept bank, yielding a 6% accuracy gain and a 30% gain on an LLM-based interpretability metric.

  3. Towards Interpretable PolSAR Image Classification: Polarimetric Scattering Mechanism Informed Concept Bottleneck and Kolmogorov-Arnold Network

    eess.IV 2025-07 conditional novelty 5.0 of 10

    A concept bottleneck model built from polarimetric target decomposition plus a Kolmogorov-Arnold Network gives PolSAR classification with human-auditable concept predictions and symbolic decision formulas.

  4. Attributing Data for Sharpness-Aware Minimization

    cs.LG 2025-07 reject novelty 4.0 of 10

    SAM-HIF and SAM-GIF are proposed as data attribution scores for SAM-trained models, but SAM-GIF is TracIn with SAM gradients and SAM-HIF's derivation contains a load-bearing error.

  5. Stable Vision Concept Transformers for Medical Diagnosis

    cs.CV 2025-06 reject novelty 4.0 of 10

    A vision transformer with a concept bottleneck and denoised diffusion smoothing is claimed to give stable concept explanations under input perturbations while keeping diagnostic accuracy.

  6. A Comprehensive Survey on the Risks and Limitations of Concept-based Models

    cs.LG 2025-05 conditional novelty 4.0 of 10

    A survey cataloging the main vulnerabilities of supervised and unsupervised concept-based models, including concept leakage, spurious correlations, and intervention failures.

Pith tools