Provable Adversarial Robustness in In-Context Learning
Fuente:
arXiv
Enregistré dans:
| Auteur principal: | |
|---|---|
| Format: | Preprint |
| Publié: |
2026
|
| Sujets: | |
| Accès en ligne: | |
| Tags: |
Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
|
| _version_ | 1866908842488496128 |
|---|---|
| author | Zhang, Di |
| author_facet | Zhang, Di |
| contents | Large language models adapt to new tasks through in-context learning (ICL) without parameter updates. Current theoretical explanations for this capability assume test tasks are drawn from a distribution similar to that seen during pretraining. This assumption overlooks adversarial distribution shifts that threaten real-world reliability. To address this gap, we introduce a distributionally robust meta-learning framework that provides worst-case performance guarantees for ICL under Wasserstein-based distribution shifts. Focusing on linear self-attention Transformers, we derive a non-asymptotic bound linking adversarial perturbation strength ($ρ$), model capacity ($m$), and the number of in-context examples ($N$). The analysis reveals that model robustness scales with the square root of its capacity ($ρ_{\text{max}} \propto \sqrt{m}$), while adversarial settings impose a sample complexity penalty proportional to the square of the perturbation magnitude ($N_ρ- N_0 \propto ρ^2$). Experiments on synthetic tasks confirm these scaling laws. These findings advance the theoretical understanding of ICL's limits under adversarial conditions and suggest that model capacity serves as a fundamental resource for distributional robustness. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2602_17743 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | Provable Adversarial Robustness in In-Context Learning Zhang, Di Machine Learning 62G35, 62M20, 68T07, 62G08 I.2.6; G.3; F.2.1 Large language models adapt to new tasks through in-context learning (ICL) without parameter updates. Current theoretical explanations for this capability assume test tasks are drawn from a distribution similar to that seen during pretraining. This assumption overlooks adversarial distribution shifts that threaten real-world reliability. To address this gap, we introduce a distributionally robust meta-learning framework that provides worst-case performance guarantees for ICL under Wasserstein-based distribution shifts. Focusing on linear self-attention Transformers, we derive a non-asymptotic bound linking adversarial perturbation strength ($ρ$), model capacity ($m$), and the number of in-context examples ($N$). The analysis reveals that model robustness scales with the square root of its capacity ($ρ_{\text{max}} \propto \sqrt{m}$), while adversarial settings impose a sample complexity penalty proportional to the square of the perturbation magnitude ($N_ρ- N_0 \propto ρ^2$). Experiments on synthetic tasks confirm these scaling laws. These findings advance the theoretical understanding of ICL's limits under adversarial conditions and suggest that model capacity serves as a fundamental resource for distributional robustness. |
| title | Provable Adversarial Robustness in In-Context Learning |
| topic | Machine Learning 62G35, 62M20, 68T07, 62G08 I.2.6; G.3; F.2.1 |
| url | https://arxiv.org/abs/2602.17743 |