← Back to all news
arXiv AI By Marek Masiak, Andrzej Szablewski, Nicola Cancedda, Constantin Venhoff

Interpretable sparse fine-tuning with TopKLoRA enables faithful circuit discovery

Read the original on arXiv AI →

The paper introduces TopKLoRA, an interpretable parameter‑efficient fine‑tuning method that activates only the top‑k adapter latents per token. Using this sparse LoRA, the authors fine‑tune large language models on a poisoned dataset, identify and ablate the minimal circuit responsible for a backdoor, and show that the resulting circuits are an order of magnitude smaller than dense LoRA baselines. They also demonstrate how to construct model organisms with localized circuits via gradient‑routing, supporting the development of interpretable‑by‑design architectures.

Machine-generated by The Flow from the publisher's headline and feed description — not written or checked by a human. The full article lives at arXiv AI.

Related stories

arXiv AI
Oct 2

Backdoor Purification for LoRA-Tuned LLMs via Null-Space Projection

The paper introduces a method to purify LoRA-tuned large language models (LLMs) against backdoor attacks without relying on trigger knowledge, clean references, or retraining. By extracting high‑fidelity backdoor directions and projecting LoRA updates onto orthogonal null spaces in input and output channels, the approach reduces attack success rates from nearly 100% to under 10%. Experiments demonstrate that this null‑space projection preserves both the base model’s general capabilities and the new downstream skills learned through the adapter across various tasks.

By Jianwei Li, Jung-Eun Kim
llmsnlpfine-tuningmultimodalbenchmarks
More like this →
arXiv AI
Oct 2

Backdoor Containment via Expert Quarantine and Shutdown in LLMs

The paper introduces Quarantined Expert Shutdown (QES), a new backdoor containment strategy for large language models. QES allows backdoor learning to occur during training but routes it into a designated, quarantined expert that can be disabled at deployment. The method achieves significant reductions in attack success rates while largely preserving model utility.

By Jianwei Li, Min-Seon Kim, Jung-Eun Kim
llmsfine-tuningmultimodal
More like this →
arXiv Machine Learning
Jul 30

ToxScreen: Detecting Whether an LLM Has Been Poisoned

arXiv:2607. 26849v1 Announce Type: cross Abstract: As large language models (LLMs) are deployed in high-stakes domains, adversaries may poison training data to implant backdoors: hidden triggers that covertly manipulate model behavior at inference time.

By Anthony Hughes, Nicole Xing, Collin Francel, Andy Kim, Andrew Draganov
llmsbenchmarks
More like this →