arXiv Machine Learning August 5, 2026 By Nathan Labiosa, David Buff, Ena Nayak, Erica Donno

Sensitivity, Causality, and Repair Dissociate: A Layer-Wise Analysis of Perturbation Robustness and Its Scaling

Read the original on arXiv Machine Learning →

arXiv:2608. 03842v1 Announce Type: cross Abstract: When a language model fails on surface-perturbed input (typos, OCR noise, homophones), "which layer is responsible" has three natural operationalizations: where representations diverge most (sensitivity), where restoring clean activations recovers the prediction (causality), and where a small adapter can repair the damage (compensatory capacity) - and we show these three layer maps dissociate.

Summary generated by The Flow from the publisher's feed. The full article lives at arXiv Machine Learning.

arXiv Machine Learning

6d ago

Orientation, not magnitude: the causal structure of task-vector interference in merged language models

arXiv:2608. 11797v1 Announce Type: new Abstract: Model merging by task arithmetic works until it doesn't, and the field diagnoses why with magnitudes: layerwise representation bias, deviations from cross-task linearity, parameter overlap.

By Chencheng Zhu

llms efficiency safety

arXiv AI

Jul 3

Breaking Safety at the Token Boundary: How BPE Tokenization Creates Exploitable Gaps in LLM Alignment

arXiv:2607. 01239v1 Announce Type: cross Abstract: Character-level perturbations bypass safety alignment in modern LLMs despite leaving prompts human-readable.

By Tung-Ling Li, Hongliang Liu, Yuhao Wu

llms nlp reinforcement-learning fine-tuning multimodal safety

arXiv AI

Aug 7

When Self-Evolution Backfires: Pre-Commit Gating against Skill Contamination in LLM Agents

arXiv:2608. 05810v1 Announce Type: new Abstract: Self-evolving agents accumulate capability by distilling reusable skills from their execution trajectories, but we find this process is not monotonic: past a critical pool size, newly added skills degrade performance instead of improving it.

By Linfang Shang, Ming Xu, Yiding Sun, Tianle Xia, Lingxiang Hu, Lan Xu, Ning Zheng

llms agents benchmarks

arXiv Machine Learning

Aug 12

Same Targets, Different Computation: How Post-Training Divides Work Across Model Layers

arXiv:2605. 07284v2 Announce Type: replace Abstract: A late-layer change learned during post-training may work on the base model's earlier state, or it may depend on earlier computation learned with it.

By Yifan Zhou

llms fine-tuning

arXiv Machine Learning

Jun 2

Task Structure Reverses Layerwise State Encoding in Sequence Models

arXiv:2606. 00926v1 Announce Type: new Abstract: Mechanistic studies of sequence models often treat layerwise state encodings as architectural traits: recurrent models concentrate readable state, attention-based models distribute it.

By Yuhang Jiang

llms fine-tuning benchmarks

arXiv AI

Aug 6

Necessary, Decodable and Reversible, Yet Not Transferable: A Stress Test for Attention-Head Role Claims

arXiv:2606. 08292v2 Announce Type: replace Abstract: Mechanistic studies often assign a component a role when removing it damages a behavior, its activation linearly encodes task information, and restoring that activation repairs the damage.

By Philip Quirke

Related stories

Orientation, not magnitude: the causal structure of task-vector interference in merged language models

Breaking Safety at the Token Boundary: How BPE Tokenization Creates Exploitable Gaps in LLM Alignment

When Self-Evolution Backfires: Pre-Commit Gating against Skill Contamination in LLM Agents

Same Targets, Different Computation: How Post-Training Divides Work Across Model Layers

Task Structure Reverses Layerwise State Encoding in Sequence Models

Necessary, Decodable and Reversible, Yet Not Transferable: A Stress Test for Attention-Head Role Claims