arXiv Machine Learning

CircuitSteer: Geometrically Aligned Multi-Layer Steering via Sparse Autoencoder Circuits

arXiv:2608. 05732v1 Announce Type: new Abstract: Controlling the behavior of large language models (LLMs) remains a critical challenge for AI alignment.

arXiv AI

Jun 16

Scalable Circuit Learning for Interpreting Large Language Models

arXiv:2606. 16939v1 Announce Type: cross Abstract: A prominent research direction in mechanistic interpretability is learning sparse circuits over LLM components to reveal how they jointly produce model behavior.

By Naiyu Yin, Dennis Wei, Tian Gao, Amit Dhurandhar, Karthikeyan Natesan Ramamurthy, Yue Yu

llms benchmarks safety

arXiv AI

Jun 12

Language Model Circuits Are Sparse in the Neuron Basis

arXiv:2601. 22594v2 Announce Type: replace-cross Abstract: The high-level concepts that a neural network uses to perform computation need not be aligned to individual neurons (Smolensky, 1986).

By Aryaman Arora, Zhengxuan Wu, Jacob Steinhardt, Sarah Schwettmann

llms benchmarks safety

arXiv AI

Jun 26

Discovering Millions of Interpretable Features with Sparse Autoencoders

arXiv:2606. 26620v1 Announce Type: cross Abstract: Sparse autoencoders (SAEs) have emerged as a powerful tool for decomposing superposed language model representations into sparse and interpretable features.

By XinYang He, Wei Wang, Bing Zhao, Xuan Ren, WenBo Li, WeiXu Qiao, Hu Wei, Lin Qu

llms efficiency

arXiv Machine Learning

Jul 9

Distributed Sparse Interventions in Language Models

arXiv:2607. 07128v1 Announce Type: new Abstract: Language models perform a wide range of tasks at varying levels of abstraction with the capacity to flexibly infer tasks from context, execute multiple tasks simultaneously, and select among competing tasks.

By Maximilian S. Ernst (Max Planck School of Cognition, Center for Lifespan Psychology Max Planck Institute for Human Development, Machine Learning Group Technische Universit\"at Berlin), Lorenz Linhardt (Machine Learning Group Technische Universit\"at Berlin, Berlin Institute for the Foundations of Learning and Data), Aaron Peikert (Center for Lifespan Psychology Max Planck Institute for Human Development), Oliver Eberle (Machine Learning Group Technische Universit\"at Berlin, Berlin Institute for the Foundations of Learning and Data)

llms

arXiv AI

Jun 16

DifFRACT: Diffusion Feature Reconstruction and Attribution for Circuit Tracing

arXiv:2606. 15796v1 Announce Type: cross Abstract: Mechanistic interpretability seeks to explain neural network behavior by decomposing model computations into interpretable features and circuits.

By Artyom Mazur, Nina Konovalova, Aibek Alanov

llms diffusion efficiency multimodal benchmarks safety

arXiv Machine Learning

Jun 9

Beyond Linear Activation Steering: Invertible Latent Transformations for Controlling LLM Behavior

arXiv:2606. 08454v1 Announce Type: new Abstract: Activation steering provides a lightweight inference-time mechanism for controlling large language models (LLMs) by modifying their internal activation vectors toward desired behaviors.

By Tuc Nguyen, Thai Le

llms diffusion benchmarks safety

arXiv AI

Jul 13

GrAInS: Gradient-based Attribution for Inference-Time Steering of LLMs and VLMs

arXiv:2507. 18043v2 Announce Type: replace-cross Abstract: Inference-time steering methods offer a lightweight alternative to fine-tuning large language models (LLMs) and vision-language models (VLMs) by modifying internal activations at test time without updating model weights.

By Duy Nguyen, Archiki Prasad, Elias Stengel-Eskin, Mohit Bansal

llms fine-tuning multimodal safety

arXiv Machine Learning

Jun 2

Resonant Context Anchoring: Decoupling Attention Routing and Signal Gain at Inference Time

arXiv:2606. 01923v1 Announce Type: cross Abstract: Large Language Models (LLMs) frequently exhibit "contextual disregard" when faced with input evidence that conflicts with their internal parametric memory, leading to persistent factual hallucinations.

By Mingkuan Zhao, Yide Gao, Wentao Hu, Suquan Chen, Tianchen Huang, Zhenhua An, Zetao Chang, Xiayu Sun, Yuheng Min

llms efficiency safety

arXiv AI

Aug 3

Wrong Code, Right Structure: Learning Netlist Representations from Imperfect LLM-Generated RTL

arXiv:2603. 09161v2 Announce Type: replace-cross Abstract: Learning effective netlist representations is fundamentally constrained by the scarcity of labeled datasets, as real designs are protected by Intellectual Property (IP) and costly to annotate.

By Siyang Cai, Cangyuan Li, Haoyu Gao, Kun Wang, Yinhe Han, Ying Wang

llms benchmarks

arXiv AI

Jun 11

ICA Lens: Interpreting Language Models Without Training Another Dictionary

arXiv:2606. 11722v1 Announce Type: cross Abstract: Finding interpretable directions in language-model representations is critical for understanding and controlling model behavior.

By Sida Liu, Feijiang Han

llms safety

arXiv AI

Jun 26

Detecting and Controlling Sycophancy with Cascading Linear Features

arXiv:2606. 26155v1 Announce Type: new Abstract: Interpreting and controlling model behaviors through activation steering methods requires many pairs of contrastive samples that clearly exhibit desired or undesired behavior.

By Maty Bohacek, Rishub Jain, Nicholas Dufour, Thomas Leung, Chris Bregler, Roma Patel

llms safety

Hugging Face Trending Papers

6d ago

SAEVerbalizer: Generating Explanations for Sparse Autoencoder Features via Representation Verbalization

Sparse autoencoders (SAEs) are proposed to extract numerous features from large language model (LLM) representations, yet explaining these features still relies primarily on external observation. This reliance leads to superficial explanations inferred from observed model behavior and computational inefficiency from collecting such behavioral evidence at scale.

llms

Related stories