Imagen de Portada

Controllable LLM Reasoning via Sparse Autoencoder-Based Steering

Fuente: arXiv

Guardado en:

Detalles Bibliográficos
Autores principales:	Fang, Yi, Wang, Wenjie, Xue, Mingfeng, Deng, Boyi, Xu, Fengli, Liu, Dayiheng, Feng, Fuli
Formato:	Preprint
Publicado:	2026
Materias:	Artificial Intelligence Computation and Language
Acceso en línea:	Acceder al recurso
Etiquetas:	Agregar Etiqueta Sin Etiquetas, Sea el primero en etiquetar este registro!

Ejemplares similares

DotaMath: Decomposition of Thought with Code Assistance and Self-correction for Mathematical Reasoning
por: Li, Chengpeng, et al.
Publicado: (2024)

SAEMark: Steering Personalized Multilingual LLM Watermarks with Sparse Autoencoders
por: Yu, Zhuohao, et al.
Publicado: (2025)

Steering LVLMs via Sparse Autoencoder for Hallucination Mitigation
por: Hua, Zhenglin, et al.
Publicado: (2025)

SASFT: Sparse Autoencoder-guided Supervised Finetuning to Mitigate Unexpected Code-Switching in LLMs
por: Deng, Boyi, et al.
Publicado: (2025)

Mechanistic Knobs in LLMs: Retrieving and Steering High-Order Semantic Features via Sparse Autoencoders
por: Zhang, Ruikang, et al.
Publicado: (2026)

Denoising Concept Vectors with Sparse Autoencoders for Improved Language Model Steering
por: Zhao, Haiyan, et al.
Publicado: (2025)

Teaching Language Models to Reason with Tools
por: Li, Chengpeng, et al.
Publicado: (2025)

Unveiling Language-Specific Features in Large Language Models via Sparse Autoencoders
por: Deng, Boyi, et al.
Publicado: (2025)

CorrSteer: Generation-Time LLM Steering via Correlated Sparse Autoencoder Features
por: Cho, Seonglae, et al.
Publicado: (2025)

One Adapts to Any: Meta Reward Modeling for Personalized LLM Alignment
por: Cai, Hongru, et al.
Publicado: (2026)

Improving Steering Vectors by Targeting Sparse Autoencoder Features
por: Chalnev, Sviatoslav, et al.
Publicado: (2024)

Constrain Alignment with Sparse Autoencoders
por: Yin, Qingyu, et al.
Publicado: (2024)

Causal Language Control in Multilingual Transformers via Sparse Feature Steering
por: Chou, Cheng-Ting, et al.
Publicado: (2025)

CoRT: Code-integrated Reasoning within Thinking
por: Li, Chengpeng, et al.
Publicado: (2025)

TAT-LLM: A Specialized Language Model for Discrete Reasoning over Tabular and Textual Data
por: Zhu, Fengbin, et al.
Publicado: (2024)

Steering LLMs? Actually, Sparse Autoencoders can outperform simple baselines
por: Jørgensen, Mikkel Godsk, et al.
Publicado: (2026)

Enhancing LLM Steering through Sparse Autoencoder-Based Vector Refinement
por: Wang, Anyi, et al.
Publicado: (2025)

Route-and-Reason: Scaling Large Language Model Reasoning with Reinforced Model Router
por: Shao, Chenyang, et al.
Publicado: (2025)

SparseRM: A Lightweight Preference Modeling with Sparse Autoencoder
por: Liu, Dengcan, et al.
Publicado: (2025)

Less is More: Improving LLM Alignment via Preference Data Selection
por: Deng, Xun, et al.
Publicado: (2025)

HeartLLM: Discretized ECG Tokenization for LLM-Based Diagnostic Reasoning
por: Yang, Jinning, et al.
Publicado: (2025)

The Evolution of Thought: Tracking LLM Overthinking via Reasoning Dynamics Analysis
por: Wei, Zihao, et al.
Publicado: (2025)

Improving LLM Reasoning through Interpretable Role-Playing Steering
por: Wang, Anyi, et al.
Publicado: (2025)

Evaluating Mathematical Reasoning of Large Language Models: A Focus on Error Identification and Correction
por: Li, Xiaoyuan, et al.
Publicado: (2024)

Effectively Steer LLM To Follow Preference via Building Confident Directions
por: Song, Bingqing, et al.
Publicado: (2025)

SAIF: A Sparse Autoencoder Framework for Interpreting and Steering Instruction Following of Language Models
por: He, Zirui, et al.
Publicado: (2025)

Inference-time Alignment via Sparse Junction Steering
por: Hu, Runyi, et al.
Publicado: (2026)

Examining False Positives under Inference Scaling for Mathematical Reasoning
por: Wang, Yu, et al.
Publicado: (2025)

Control Reinforcement Learning: Interpretable Token-Level Steering of LLMs via Sparse Autoencoder Features
por: Cho, Seonglae, et al.
Publicado: (2026)

A Systematic Analysis of the Impact of Persona Steering on LLM Capabilities
por: Chen, Jiaqi, et al.
Publicado: (2026)

Reasoning Like a Doctor: Improving Medical Dialogue Systems via Diagnostic Reasoning Process Alignment
por: Xu, Kaishuai, et al.
Publicado: (2024)

Guiding LLM Post-training Data Engineering with Model Internals from Sparse Autoencoders
por: Jing, Yi, et al.
Publicado: (2026)

ProtSAE: Disentangling and Interpreting Protein Language Models via Semantically-Guided Sparse Autoencoders
por: Liu, Xiangyu, et al.
Publicado: (2025)

AxBench: Steering LLMs? Even Simple Baselines Outperform Sparse Autoencoders
por: Wu, Zhengxuan, et al.
Publicado: (2025)

Layer-Wise Perturbations via Sparse Autoencoders for Adversarial Text Generation
por: Shu, Huizhen, et al.
Publicado: (2025)

Token Signature: Predicting Chain-of-Thought Gains with Token Decoding Feature in Large Language Models
por: Liu, Peijie, et al.
Publicado: (2025)

AlpsBench: An LLM Personalization Benchmark for Real-Dialogue Memorization and Preference Alignment
por: Xiao, Jianfei, et al.
Publicado: (2026)

Disentangling concept semantics via multilingual averaging in Sparse Autoencoders
por: O'Reilly, Cliff, et al.
Publicado: (2025)

Graph-Regularized Sparse Autoencoders for LLM Safety Steering
por: Yeon, Jehyeok, et al.
Publicado: (2025)

Directional Attractors in LLM Reasoning: How Similarity Retrieval Steers Iterative Summarization Based Reasoning
por: Tekin, Cagatay, et al.
Publicado: (2025)