Copertina

Short-circuiting Shortcuts: Mechanistic Investigation of Shortcuts in Text Classification

Fuente: arXiv

Salvato in:

Dettagli Bibliografici
Autori principali:	Eshuijs, Leon, Wang, Shihan, Fokkens, Antske
Natura:	Preprint
Pubblicazione:	2025
Soggetti:	Machine Learning Computation and Language
Accesso online:	Acceder al recurso
Tags:	Aggiungi Tag Nessun Tag, puoi essere il primo ad aggiungerne!!

Documenti analoghi

Balancing the Scales: Reinforcement Learning for Fair Classification
di: Eshuijs, Leon, et al.
Pubblicazione: (2024)

Safety Training Modulates Harmful Misalignment Under On-Policy RL, But Direction Depends on Environment Design
di: Eshuijs, Leon, et al.
Pubblicazione: (2026)

Navigating the Shortcut Maze: A Comprehensive Analysis of Shortcut Learning in Text Classification by Language Models
di: Zhou, Yuqing, et al.
Pubblicazione: (2024)

Spurious Rewards Paradox: Mechanistically Understanding How RLVR Activates Memorization Shortcuts in LLMs
di: Yan, Lecheng, et al.
Pubblicazione: (2026)

Do LLMs Overcome Shortcut Learning? An Evaluation of Shortcut Challenges in Large Language Models
di: Yuan, Yu, et al.
Pubblicazione: (2024)

Investigating the Robustness of Modelling Decisions for Few-Shot Cross-Topic Stance Detection: A Preregistered Study
di: Reuver, Myrthe, et al.
Pubblicazione: (2024)

Reasoning in Transformers -- Mitigating Spurious Correlations and Reasoning Shortcuts
di: Enström, Daniel, et al.
Pubblicazione: (2024)

Control the Temperature: Selective Sampling for Diverse and High-Quality LLM Outputs
di: Troshin, Sergey, et al.
Pubblicazione: (2025)

Seeing to Generalize: How Visual Data Corrects Binding Shortcuts
di: Buzeta, Nicolas, et al.
Pubblicazione: (2026)

MMLU-Pro+: Evaluating Higher-Order Reasoning and Shortcut Learning in LLMs
di: Taghanaki, Saeid Asgari, et al.
Pubblicazione: (2024)

Learning Shortcuts: On the Misleading Promise of NLU in Language Models
di: Bihani, Geetanjali, et al.
Pubblicazione: (2024)

Shortcut-connected Expert Parallelism for Accelerating Mixture-of-Experts
di: Cai, Weilin, et al.
Pubblicazione: (2024)

From Shortcuts to Triggers: Backdoor Defense with Denoised PoE
di: Liu, Qin, et al.
Pubblicazione: (2023)

DefVerify: Do Hate Speech Models Reflect Their Dataset's Definition?
di: Khurana, Urja, et al.
Pubblicazione: (2024)

On the Low-Rank Parametrization of Reward Models for Controlled Language Generation
di: Troshin, Sergey, et al.
Pubblicazione: (2024)

The Reliability Paradox: Exploring How Shortcut Learning Undermines Language Model Calibration
di: Bihani, Geetanjali, et al.
Pubblicazione: (2024)

Learning from Sufficient Rationales: Analysing the Relationship Between Explanation Faithfulness and Token-level Regularisation Strategies
di: Kamp, Jonathan, et al.
Pubblicazione: (2025)

Improving Causal Interventions in Amnesic Probing with Mean Projection or LEACE
di: Dobrzeniecka, Alicja, et al.
Pubblicazione: (2025)

Not Eliminate but Aggregate: Post-Hoc Control over Mixture-of-Experts to Address Shortcut Shifts in Natural Language Understanding
di: Honda, Ukyo, et al.
Pubblicazione: (2024)

The Role of Syntactic Span Preferences in Post-Hoc Explanation Disagreement
di: Kamp, Jonathan, et al.
Pubblicazione: (2024)

Adaptive Large Language Models By Layerwise Attention Shortcuts
di: Verma, Prateek, et al.
Pubblicazione: (2024)

CURE: Controlled Unlearning for Robust Embeddings -- Mitigating Conceptual Shortcuts in Pre-Trained Language Models
di: Kocak, Aysenur, et al.
Pubblicazione: (2025)

Investigating the Influence of Prompt-Specific Shortcuts in AI Generated Text Detection
di: Park, Choonghyun, et al.
Pubblicazione: (2024)

On the Foundations of Shortcut Learning
di: Hermann, Katherine L., et al.
Pubblicazione: (2023)

Asking a Language Model for Diverse Responses
di: Troshin, Sergey, et al.
Pubblicazione: (2025)

Crowd-Calibrator: Can Annotator Disagreement Inform Calibration in Subjective Tasks?
di: Khurana, Urja, et al.
Pubblicazione: (2024)

Models Know Their Shortcuts: Deployment-Time Shortcut Mitigation
di: Li, Jiayi, et al.
Pubblicazione: (2026)

Persian Slang Text Conversion to Formal and Deep Learning of Persian Short Texts on Social Media for Sentiment Classification
di: Khazeni, Mohsen, et al.
Pubblicazione: (2024)

ShortcutProbe: Probing Prediction Shortcuts for Learning Robust Models
di: Zheng, Guangtao, et al.
Pubblicazione: (2025)

Shortcut Learning Susceptibility in Vision Classifiers
di: Suhail, Pirzada, et al.
Pubblicazione: (2025)

L3Cube-MahaNews: News-based Short Text and Long Document Classification Datasets in Marathi
di: Mittal, Saloni, et al.
Pubblicazione: (2024)

L3Cube-IndicNews: News-based Short Text and Long Document Classification Datasets in Indic Languages
di: Mirashi, Aishwarya, et al.
Pubblicazione: (2024)

Shortcuts Arising from Contrast: Effective and Covert Clean-Label Attacks in Prompt-Based Learning
di: Xie, Xiaopeng, et al.
Pubblicazione: (2024)

One Step Diffusion via Shortcut Models
di: Frans, Kevin, et al.
Pubblicazione: (2024)

The Readout Shortcut: Positional Number Copying Dominates Arithmetic CoT Readout in Small Language Models
di: Liu, Ming
Pubblicazione: (2026)

Mechanistic origins of catastrophic forgetting: why RL preserves circuits better than SFT?
di: Nunez, Jeanmely Rojas, et al.
Pubblicazione: (2026)

Generative Classifiers Avoid Shortcut Solutions
di: Li, Alexander C., et al.
Pubblicazione: (2025)

Bias in the Shadows: Explore Shortcuts in Encrypted Network Traffic Classification
di: Wang, Chuyi, et al.
Pubblicazione: (2026)

Stochastic Adversarial Networks for Multi-Domain Text Classification
di: Wang, Xu, et al.
Pubblicazione: (2024)

Understand the Effectiveness of Shortcuts through the Lens of DCA
di: Sun, Youran, et al.
Pubblicazione: (2024)