Copertina

Spectral Attention Steering for Prompt Highlighting

Fuente: arXiv

Salvato in:

Dettagli Bibliografici
Autori principali:	Li, Weixian Waylon, Niu, Yuchen, Yang, Yongxin, Li, Keshuang, Ma, Tiejun, Cohen, Shay B.
Natura:	Preprint
Pubblicazione:	2026
Soggetti:	Computation and Language Artificial Intelligence
Accesso online:	Acceder al recurso
Tags:	Aggiungi Tag Nessun Tag, puoi essere il primo ad aggiungerne!!

Documenti analoghi

Time is Not a Label: Continuous Phase Rotation for Temporal Knowledge Graphs and Agentic Memory
di: Li, Weixian Waylon, et al.
Pubblicazione: (2026)

TSPRank: Bridging Pairwise and Listwise Methods with a Bilinear Travelling Salesman Model
di: Li, Weixian Waylon, et al.
Pubblicazione: (2024)

Summoning the Oracle to Slay It: Mitigating Look-Ahead Bias in Financial Backtesting with Large Language Models
di: Li, Weixian Waylon, et al.
Pubblicazione: (2026)

Modeling News Interactions and Influence for Financial Market Prediction
di: Wang, Mengyu, et al.
Pubblicazione: (2024)

Learn to Rank Risky Investors: A Case Study of Predicting Retail Traders' Behaviour and Profitability
di: Li, Weixian Waylon, et al.
Pubblicazione: (2025)

Self-Improving World Modelling with Latent Actions
di: Qiu, Yifu, et al.
Pubblicazione: (2026)

Interpreting Context Look-ups in Transformers: Investigating Attention-MLP Interactions
di: Neo, Clement, et al.
Pubblicazione: (2024)

Prompt-Activation Duality: Improving Activation Steering via Attention-Level Interventions
di: Kang, Diancheng, et al.
Pubblicazione: (2026)

Can Large Language Models Follow Concept Annotation Guidelines? A Case Study on Scientific and Financial Domains
di: Fonseca, Marcio, et al.
Pubblicazione: (2023)

Can Large Language Model Summarizers Adapt to Diverse Scientific Communication Goals?
di: Fonseca, Marcio, et al.
Pubblicazione: (2024)

Self Knowledge Re-expression: A Fully Local Method for Adapting LLMs to Tasks Using Intrinsic Knowledge
di: Wang, Mengyu, et al.
Pubblicazione: (2026)

`Keep it Together': Enforcing Cohesion in Extractive Summaries by Simulating Human Memory
di: Cardenas, Ronald, et al.
Pubblicazione: (2024)

Spectral Editing of Activations for Large Language Model Alignment
di: Qiu, Yifu, et al.
Pubblicazione: (2024)

Think While You Write: Hypothesis Verification Promotes Faithful Knowledge-to-Text Generation
di: Qiu, Yifu, et al.
Pubblicazione: (2023)

Fusion Steering: Prompt-Specific Activation Control
di: Chang, Waldemar, et al.
Pubblicazione: (2025)

Steer Like the LLM: Activation Steering that Mimics Prompting
di: Heyman, Geert, et al.
Pubblicazione: (2026)

Steering When Necessary: Flexible Steering Large Language Models with Backtracking
di: Cheng, Zifeng, et al.
Pubblicazione: (2025)

Prompt-Based Value Steering of Large Language Models
di: Abbo, Giulio Antonio, et al.
Pubblicazione: (2025)

EasySteer: A Unified Framework for High-Performance and Extensible LLM Steering
di: Xu, Haolei, et al.
Pubblicazione: (2025)

Hallucination reduction with CASAL: Contrastive Activation Steering For Amortized Learning
di: Wannan, et al.
Pubblicazione: (2025)

Steering LLM Thinking with Budget Guidance
di: Li, Junyan, et al.
Pubblicazione: (2025)

Learning Evidence Highlighting for Frozen LLMs
di: Li, Shaoang, et al.
Pubblicazione: (2026)

An Extensive Evaluation of PDDL Capabilities in off-the-shelf LLMs
di: Vyas, Kaustubh, et al.
Pubblicazione: (2025)

Beyond Prompt: Fine-grained Simulation of Cognitively Impaired Standardized Patients via Stochastic Steering
di: Zhang, Weikang, et al.
Pubblicazione: (2026)

What can Large Language Models Capture about Code Functional Equivalence?
di: Maveli, Nickil, et al.
Pubblicazione: (2024)

Magpie: Alignment Data Synthesis from Scratch by Prompting Aligned LLMs with Nothing
di: Xu, Zhangchen, et al.
Pubblicazione: (2024)

Look Before You Leap: Enhancing Attention and Vigilance Regarding Harmful Content with GuidelineLLM
di: Zhang, Shaoqing, et al.
Pubblicazione: (2024)

Salient Information Prompting to Steer Content in Prompt-based Abstractive Summarization
di: Xu, Lei, et al.
Pubblicazione: (2024)

Model Tells Itself Where to Attend: Faithfulness Meets Automatic Attention Steering
di: Zhang, Qingru, et al.
Pubblicazione: (2024)

CoMAT: Chain of Mathematically Annotated Thought Improves Mathematical Reasoning
di: Leang, Joshua Ong Jun, et al.
Pubblicazione: (2024)

Spectral Filters, Dark Signals, and Attention Sinks
di: Cancedda, Nicola
Pubblicazione: (2024)

Exploring the Personality Traits of LLMs through Latent Features Steering
di: Yang, Shu, et al.
Pubblicazione: (2024)

CogSteer: Cognition-Inspired Selective Layer Intervention for Efficiently Steering Large Language Models
di: Wang, Xintong, et al.
Pubblicazione: (2024)

CoSToM:Causal-oriented Steering for Intrinsic Theory-of-Mind Alignment in Large Language Models
di: Li, Mengfan, et al.
Pubblicazione: (2026)

ThinkPilot: Steering Reasoning Models via Automated Think-prefixes Optimization
di: Li, Sunzhu, et al.
Pubblicazione: (2025)

Self-Steering Language Models
di: Grand, Gabriel, et al.
Pubblicazione: (2025)

Coarse-to-Fine Highlighting: Reducing Knowledge Hallucination in Large Language Models
di: Lv, Qitan, et al.
Pubblicazione: (2024)

AttentionDefense: Leveraging System Prompt Attention for Explainable Defense Against Novel Jailbreaks
di: Siska, Charlotte, et al.
Pubblicazione: (2025)

SafeSteer: Localized On-Policy Distillation for Efficient Safety Alignment
di: Li, Hao, et al.
Pubblicazione: (2026)

Lost in Space? Vision-Language Models Struggle with Relative Camera Pose Estimation
di: Deng, Ken, et al.
Pubblicazione: (2026)