Image de couverture de livre

Inference-time Alignment via Sparse Junction Steering

Fuente: arXiv

Enregistré dans:

Détails bibliographiques
Auteurs principaux:	Hu, Runyi, Zhang, Jie, Zhao, Shiqian, Meng, Jiale, Li, Jiwei, Zeng, Jason, Wu, Ming, Heinrich, Michael, Wen, Yonggang, Zhang, Tianwei
Format:	Preprint
Publié:	2026
Sujets:	Computation and Language Artificial Intelligence
Accès en ligne:	Acceder al recurso
Tags:	Ajouter un tag Pas de tags, Soyez le premier à ajouter un tag!

Documents similaires

Mask Image Watermarking
par: Hu, Runyi, et autres
Publié: (2025)

Unifying Watermarking via Dimension-Aware Mapping
par: Meng, Jiale, et autres
Publié: (2026)

TRACE: Structure-Aware Character Encoding for Robust and Generalizable Document Watermarking
par: Meng, Jiale, et autres
Publié: (2026)

Robust-Wide: Robust Watermarking against Instruction-driven Image Editing
par: Hu, Runyi, et autres
Publié: (2024)

VideoShield: Regulating Diffusion-based Video Generation Models via Watermarking
par: Hu, Runyi, et autres
Publié: (2025)

SuperMark: Robust and Training-free Image Watermarking via Diffusion-based Super-Resolution
par: Hu, Runyi, et autres
Publié: (2024)

SAME: Sample Reconstruction against Model Extraction Attacks
par: Xie, Yi, et autres
Publié: (2023)

Reinforcement Learning Enhanced LLMs: A Survey
par: Wang, Shuhe, et autres
Publié: (2024)

TorchGT: A Holistic System for Large-scale Graph Transformer Training
par: Zhang, Meng, et autres
Publié: (2024)

When Memory Becomes a Vulnerability: Towards Multi-turn Jailbreak Attacks against Text-to-Image Generation Systems
par: Zhao, Shiqian, et autres
Publié: (2025)

InfraMind: A Novel Exploration-based GUI Agentic Framework for Mission-critical Industrial Management
par: Lin, Liangtao, et autres
Publié: (2025)

DSB: Dynamic Sliding Block Scheduling for Diffusion LLMs
par: Luo, Lizhuo, et autres
Publié: (2026)

SOPRAG: Multi-view Graph Experts Retrieval for Industrial Standard Operating Procedures
par: Lin, Liangtao, et autres
Publié: (2026)

Instruction Tuning for Large Language Models: A Survey
par: Zhang, Shengyu, et autres
Publié: (2023)

LAFR: Efficient Diffusion-based Blind Face Restoration via Latent Codebook Alignment Adapter
par: Li, Runyi, et autres
Publié: (2025)

Visible Yet Unreadable: A Systematic Blind Spot of Vision Language Models Across Writing Systems
par: Zhang, Jie, et autres
Publié: (2025)

Picky LLMs and Unreliable RMs: An Empirical Study on Safety Alignment after Instruction Tuning
par: Li, Guanlin, et autres
Publié: (2025)

Mind Your HEARTBEAT! Claw Background Execution Inherently Enables Silent Memory Pollution
par: Zhang, Yechao, et autres
Publié: (2026)

Three Minds, One Legend: Jailbreak Large Reasoning Model with Adaptive Stacked Ciphers
par: Nguyen, Viet-Anh, et autres
Publié: (2025)

SPPO:Efficient Long-sequence LLM Training via Adaptive Sequence Pipeline Parallel Offloading
par: Chen, Qiaoling, et autres
Publié: (2025)

Mind the Cost of Scaffold! Benign Clients May Even Become Accomplices of Backdoor Attack
par: Han, Xingshuo, et autres
Publié: (2024)

Efficient Fuzzy Private Set Intersection from Secret-shared OPRF
par: Yang, Xinpeng, et autres
Publié: (2026)

Rethinking Key-Value Cache Compression Techniques for Large Language Model Serving
par: Gao, Wei, et autres
Publié: (2025)

PromptTuner: SLO-Aware Elastic System for LLM Prompt Tuning
par: Gao, Wei, et autres
Publié: (2026)

CONCUR: High-Throughput Agentic Batch Inference of LLM via Congestion-Based Concurrency Control
par: Chen, Qiaoling, et autres
Publié: (2026)

SparseDVFS: Sparse-Aware DVFS for Energy-Efficient Edge Inference
par: Zhang, Ziyang, et autres
Publié: (2026)

Iterative Inference-time Scaling with Adaptive Frequency Steering for Image Super-Resolution
par: Zhang, Hexin, et autres
Publié: (2025)

Exploring Diverse Generation Paths via Inference-time Stiefel Activation Steering
par: Zhu, Dongxuan, et autres
Publié: (2026)

Robust and Reversible Thermal/Electro‐Responsive Supramolecular Polymeric Adhesives via Synergistic Hydrogen‐Bonds and Ionic Junctions
par: Qizhe Deng, et autres
Publié: (2024)

Robust and Reversible Thermal/Electro‐Responsive Supramolecular Polymeric Adhesives via Synergistic Hydrogen‐Bonds and Ionic Junctions
par: Qizhe Deng, et autres
Publié: (2024)

CorrSteer: Generation-Time LLM Steering via Correlated Sparse Autoencoder Features
par: Cho, Seonglae, et autres
Publié: (2025)

The Anatomy of Alignment: Decomposing Preference Optimization by Steering Sparse Features
par: Ferrao, Jeremias, et autres
Publié: (2025)

Interpretable LLM Guardrails via Sparse Representation Steering
par: He, Zeqing, et autres
Publié: (2025)

Semantic-Aware Scheduling for GPU Clusters with Large Language Models
par: Wang, Zerui, et autres
Publié: (2025)

InferDPT: Privacy-Preserving Inference for Closed-box Large Language Model
par: Tong, Meng, et autres
Publié: (2023)

TouchGuide: Inference-Time Steering of Visuomotor Policies via Touch Guidance
par: Zhang, Zhemeng, et autres
Publié: (2026)

FaithSteer-BENCH: A Deployment-Aligned Stress-Testing Benchmark for Inference-Time Steering
par: Ding, Zikang, et autres
Publié: (2026)

SteerRM: Debiasing Reward Models via Sparse Autoencoders
par: Sun, Mengyuan, et autres
Publié: (2026)

Energy-Efficient Thermal Comfort Control in Smart Buildings via Deep Reinforcement Learning
par: Gao, Guanyu, et autres
Publié: (2019)

Latent Embedding Adaptation for Human Preference Alignment in Diffusion Planners
par: Ng, Wen Zheng Terence, et autres
Publié: (2025)