conftrace_

steering vector

33 papers

Explore in graph

Also known as

SV

Co-occurring keywords

large language model (13587) sparse autoencoder (112) language model (4599) activation steering (45) model steering (26) model interpretability (519) chain-of-thought reasoning (551) activation editing (6) mvdr beamforming (3) mvdr beamformer (3)

Papers

Query-Routed Activation Editing with Truth-hierarchical Preference Optimization AAAI 2026
Do Clinical Question Answering Systems Really Need Specialised Medical Fine Tuning? EACL 2026
Shifting Perspectives: Steering Vectors for Robust Bias Mitigation in LLMs EACL 2026
Textual Steering Vectors Can Improve Visual Understanding in Multimodal Large Language Models ACL 2026
Beyond Multiple Choice: Evaluating Steering Vectors for Summarization EACL 2026
REFLEX: Self-Refining Explainable Fact-Checking via Verdict-Anchored Style Control ACL 2026
Model Whisper: Steering Vectors Unlock Large Language Models’ Potential in Test-Time AAAI 2026
PrivSV: Differentially Private Steering Vector for Large Language Models AAAI 2026
EnSToM: Enhancing Dialogue Systems with Entropy-Scaled Steering Vectors for Topic Maintenance ACL 2025
Feature Extraction and Steering for Enhanced Chain-of-Thought Reasoning in Language Models EMNLP 2025
SAE-SSV: Supervised Steering in Sparse Representation Spaces for Reliable Control of Language Models EMNLP 2025
Improving LLM Reasoning through Interpretable Role-Playing Steering EMNLP 2025
Steering Away from Harm: An Adaptive Approach to Defending Vision Language Model Against Jailbreaks CVPR 2025
Towards a Unified Paradigm of Concept Editing in Large Language Models EMNLP 2025
Understanding and Mitigating Overrefusal in LLMs from an Unveiling Perspective of Safety Decision Boundary EMNLP 2025
Unsupervised Concept Vector Extraction for Bias Control in LLMs EMNLP 2025
EasyEdit2: An Easy-to-use Steering Framework for Editing Large Language Models EMNLP 2025
Steering LLM Reasoning Through Bias-Only Adaptation EMNLP 2025
Dialz: A Python Toolkit for Steering Vectors ACL 2025
Unveiling Language-Specific Features in Large Language Models via Sparse Autoencoders ACL 2025
Cross-model Transferability among Large Language Models on the Platonic Representations of Concepts ACL 2025
Householder Pseudo-Rotation: A Novel Approach to Activation Editing in LLMs with Direction-Magnitude Perspective EMNLP 2024
InferAligner: Inference-Time Alignment for Harmlessness through Cross-Model Guidance EMNLP 2024
Word Embeddings Are Steers for Language Models ACL 2024
Steering Llama 2 via Contrastive Activation Addition ACL 2024