arXiv Machine Learning June 9, 2026 By Jiayi Li, Shijie Tang, G\"un Kaynar, Shiyi Du, Carl Kingsford

Models Know Their Shortcuts: Deployment-Time Shortcut Mitigation

Read the original on arXiv Machine Learning →

arXiv:2604. 12277v2 Announce Type: replace Abstract: Pretrained text encoders are prone to shortcut learning, relying on token-label correlations that fail once the distribution shifts in deployment.

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 Machine Learning.

safety

arXiv Machine Learning

Sep 22

ESLM: Risk-Averse Selective Language Modeling for Efficient Pretraining

arXiv:2505.19893v2 Announce Type: replace Abstract: Large language model pretraining is compute-intensive, yet many tokens contribute marginally to learning, resulting in inefficiency. We introduce E...

By Melis Ilayda Bal, Volkan Cevher, Michael Muehlebach

llms efficiency

arXiv AI

Aug 3

TextCloak: Thwarting Unauthorized LLM Exploitation via RL-Driven Unlearnable Text

arXiv:2607. 28862v1 Announce Type: cross Abstract: The rapid development of Large Language Models (LLMs) has led to significant advances across a wide range of language tasks, while simultaneously raising growing concerns about unauthorized data exploitation and privacy leakage.

By Chengshuai Zhao, Pingchuan Ma, Dawei Li, Bohan Jiang, Zhiyuan Yu, Zhen Tan, Huan Liu

llms nlp reinforcement-learning fine-tuning benchmarks safety

arXiv Machine Learning

Jun 4

Data Attribution in Large Language Models via Bidirectional Gradient Optimization

arXiv:2606. 04928v1 Announce Type: new Abstract: Large Language Models (LLMs) are increasingly deployed across diverse applications, raising critical questions for governance, accountability, and data provenance.

By Fr\'ed\'eric Berdoz, Luca A. Lanzend\"orfer, Kaan Bayraktar, Roger Wattenhofer

llms safety

arXiv AI

Aug 25

Deep Contrastive Unlearning for Language Models

Deep Contrastive Unlearning for Language Models (DeepCUT) is a framework that removes information from fine‑tuned language models by directly optimizing their latent space. It addresses the challenge of machine unlearning in black‑box models, which has been largely overlooked by previous work that only mitigated output effects. Experiments on real‑world datasets show that DeepCUT consistently outperforms baseline methods in both effectiveness and efficiency.

By Estrid He, Tabinda Sarwar, Ibrahim Khalil, Xun Yi, Ke Wang

llms diffusion fine-tuning safety

arXiv Machine Learning

Jun 4

STRIDE: Training Data Attribution via Sparse Recovery from Subset Perturbations

arXiv:2606. 05165v1 Announce Type: new Abstract: Training Data Attribution (TDA) seeks to trace a model's predictions back to its training data.

By Rishit Dagli, Abir Harrasse, Luke Zhang, Florent Draye, Amirali Abdullah, Bernhard Sch\"olkopf, Zhijing Jin

llms benchmarks

arXiv Machine Learning

Sep 22

Logits are All We Need to Adapt Closed Models

arXiv:2502.06806v5 Announce Type: replace Abstract: Many commercial Large Language Models (LLMs) are often closed-source, limiting developers to prompt tuning for aligning content generation with spe...

By Gaurush Hiranandani, Haolun Wu, Subhojyoti Mukherjee, Sanmi Koyejo

llms

One email a morning, machine-written

Related stories

ESLM: Risk-Averse Selective Language Modeling for Efficient Pretraining

TextCloak: Thwarting Unauthorized LLM Exploitation via RL-Driven Unlearnable Text

Data Attribution in Large Language Models via Bidirectional Gradient Optimization

Deep Contrastive Unlearning for Language Models

STRIDE: Training Data Attribution via Sparse Recovery from Subset Perturbations

Logits are All We Need to Adapt Closed Models