arXiv Machine Learning

Target-Token Asymmetry Can Masquerade As Depth Specialization

arXiv Machine Learning

Jul 24

Are Single-Token Sparse Autoencoder Features Causally Necessary? Layer-Depth and SAE-Family Effects

arXiv:2607. 20596v1 Announce Type: new Abstract: Sparse autoencoder (SAE) features are used to interpret and steer large language models, yet whether a feature's causal role is stable across SAE families remains untested.

By Seonglae Cho, Zekun Wu, Kleyton Da Costa, Rishi Kalra, Ilham Wicaksono, Adriano Koshiyama

llms benchmarks safety

arXiv Machine Learning

Jun 16

Emergent retokenization symmetry in large language models: phenomenology and applications

arXiv:2606. 15521v1 Announce Type: cross Abstract: Tokenization introduces representational redundancy: under a fixed token vocabulary, every byte string admits many valid token encodings, or segmentations, that decode to the same surface string.

By Kanishk Jain, Matthew Day, Tankut Can

llms computer-vision nlp benchmarks

arXiv AI

Jun 8

Endogenous Resistance to Activation Steering in Language Models

arXiv:2602. 06941v2 Announce Type: replace-cross Abstract: Large language models can recover mid-generation from task-misaligned activation steering, producing explicit verbal restarts (e.

By Alex McKenzie, Keenan Pepper, Stijn Servaes, Martin Leitgab, Murat Cubuktepe, Mike Vaiana, Diogo de Lucena, Judd Rosenblatt, Michael S. A. Graziano

llms robotics fine-tuning safety

arXiv Computation and Language

Sep 28

Where a Model Sends Its Own Repeated Token

arXiv:2609. 31181v1 Announce Type: new Abstract: Black-box model identification works by scoring a model's response to natural-language prompts.

By Nicol\'as Vera Z\'u\~niga

nlp robotics

arXiv Computation and Language

Oct 22

A Theoretical Analysis of Why Masked Diffusion Models Mitigate the Reversal Curse

arXiv:2602.02133v3 Announce Type: replace-cross Abstract: Autoregressive language models (ARMs) suffer from the reversal curse: after learning ''$A$ is $B$,'' they often fail on the reverse query ''$...

By Moongyu Jeon, Sangwoo Shin, BumJun Kim, Kyelim Lee, Albert No

llms diffusion

arXiv AI

Jul 24

Attention Degradation, Function Token Anchoring, and the Limits of Attention-Based Intervention in Large Language Models

arXiv:2607. 20524v1 Announce Type: new Abstract: Mean cross-positional attention degradation is widely reported in transformer interpretability, yet whether it causally limits contextual retrieval remains untested.

By Sagar Dangal, Manoj Shakya

llms safety

arXiv AI

Oct 15

From Relation-Specific Heads to Shared Interventions: Contextual Entrainment in Language Models

arXiv:2610.18674v1 Announce Type: cross Abstract: Large language models rely on prompt context to perform in-context learning, follow formats, and adapt to local evidence. This ability, however, also...

By Jinghan Zheng, Jiajun Shi

llms benchmarks

arXiv Computation and Language

Sep 15

Data Attribution of Emergent Misalignment with Persona Features

arXiv:2608.11025v2 Announce Type: replace Abstract: Emergent misalignment (EM) is the phenomenon where fine-tuning a language model on a narrow task leads to harmful behavior in unrelated domains. A...

By Clemens Vetter, David Kacz\'er, Lucie Flek, Florian Mai

llms robotics fine-tuning safety

arXiv AI

Jun 12

Prefill Awareness in Large Language Models

arXiv:2606. 12747v1 Announce Type: new Abstract: Safety-relevant studies of language models, including alignment and jailbreaking evaluations and AI control protocols, often rely on prefilling model outputs.

By Andy Wang, Parv Mahajan, David Demitri Africa, Alexandra Souly, Jordan Taylor, Robert Kirk

llms agents benchmarks safety

arXiv AI

Sep 2

Attention Sensitivity Is Not Enough: Dissociating Attention-Level and Behavioural In-Context Learning under Fine-Tuning

arXiv:2609.00064v1 Announce Type: cross Abstract: In-context learning (ICL) lets large language models adapt to new tasks from demonstrations, and fine-tuning can erode this behaviour. Many preservat...

By Jinyuan Zhang, Peng He, He Hu, Yin Yuan, ShengShuo Jiao

llms fine-tuning benchmarks

arXiv AI

Jul 17

In-Place Tokenizer Expansion for Pre-trained LLMs

arXiv:2607. 15232v1 Announce Type: cross Abstract: A tokenizer fixed at the start of pre-training allocates vocabulary in proportion to the pre-training corpus, reflecting the deployment priorities at that time.

By Jimmy T. H. Smith, Tarek Dakhran, Alberto Cabrera, Simon S. Lee, Paul Pak, Aditya Tadimeti, Tim Seyde, Maxime Labonne, Alexander Amini, Mathias Lechner

llms rag nlp

arXiv AI

Aug 12

Attention-Path Fragility as an Uncertainty Signal in Large Language Models

arXiv:2608. 11138v1 Announce Type: cross Abstract: We propose that a model's uncertainty about a token is reflected not only in the breadth of its output distribution but also in whether a confident prediction is \emph{fragile} under perturbation of its attention pathways.

By Minsoo Kim, Sungyoung Ji, Kisung Moon, Ilyong Yoon

llms benchmarks

Related stories