method · first seen 4 Oct 2026
gradient-based attribution
Also: gradient attribution, gradient-based data attribution
Attributing a model's prediction to influential internal states using gradients of the output.
Mentions per week
PapersCodeDiscussion
Earliest sources
- ScopeSAE: Model-Scope Feature Discovery with Interpretable Layer SelectionPapers · 4 Oct 2026 · Qingwen Zeng, Zehao Fu, Shuyu Meng, Linghan Huang, Jiayi Zhang, Chenglin Wu, Lin…
- Kernelized Parameter Sensitivity Atlases for Amortized Conditional InterpretabilityPapers · 8 Oct 2026 · Davide Wiest, Ahanaf Hasan Ariq, Abdul Monaf Chowdhury
- Functional Subnetwork Localization for Efficient Continual Learning in Large Language ModelsPapers · 15 Oct 2026 · Shuoran Jiang, Wentao Zhu, Yang Xiang, Dongfeng Li, Yunan Zhang, Mengchen Zhao,…
- Robust LLM Data Selection via Content-Invariant Gradient AttributionPapers · 27 Oct 2026 · Jia Gu, Liang Pang, Huawei Shen
- Follow the Gradient: Tracing Safety Degradation Back to Training DataPapers · 27 Oct 2026 · Saket Reddy, Amro Abbas, Anshuman Suri, Ari Morcos, David Schwab, Pratyush Maini
- Same Function, Different Gradients: Evaluating Gradient Reliability under ReparameterizationPapers · 1 Nov 2026 · Bum Jun Kim
Latest sources
- Deep Sequence Model for Genome Wide Discovery of Coding and Regulatory Element SignaturesPapers · 15 Nov 2026 · Rayhaneh Shabani Nia, Ali Karkehabadi
- Same Function, Different Gradients: Evaluating Gradient Reliability under ReparameterizationPapers · 1 Nov 2026 · Bum Jun Kim
- Follow the Gradient: Tracing Safety Degradation Back to Training DataPapers · 27 Oct 2026 · Saket Reddy, Amro Abbas, Anshuman Suri, Ari Morcos, David Schwab, Pratyush Maini
- Robust LLM Data Selection via Content-Invariant Gradient AttributionPapers · 27 Oct 2026 · Jia Gu, Liang Pang, Huawei Shen
- Functional Subnetwork Localization for Efficient Continual Learning in Large Language ModelsPapers · 15 Oct 2026 · Shuoran Jiang, Wentao Zhu, Yang Xiang, Dongfeng Li, Yunan Zhang, Mengchen Zhao,…
- Kernelized Parameter Sensitivity Atlases for Amortized Conditional InterpretabilityPapers · 8 Oct 2026 · Davide Wiest, Ahanaf Hasan Ariq, Abdul Monaf Chowdhury
- ScopeSAE: Model-Scope Feature Discovery with Interpretable Layer SelectionPapers · 4 Oct 2026 · Qingwen Zeng, Zehao Fu, Shuyu Meng, Linghan Huang, Jiayi Zhang, Chenglin Wu, Lin…