Copertina

Predicting Where Steering Vectors Succeed

Fuente: arXiv

Salvato in:

Dettagli Bibliografici
Autore principale:	Billa, Jayadev
Natura:	Preprint
Pubblicazione:	2026
Soggetti:	Machine Learning Computation and Language
Accesso online:	Acceder al recurso
Tags:	Aggiungi Tag Nessun Tag, puoi essere il primo ad aggiungerne!!

Documenti analoghi

Modality Collapse as Mismatched Decoding: Information-Theoretic Limits of Multimodal LLMs
di: Billa, Jayadev
Pubblicazione: (2026)

The Geometric Anatomy of Capability Acquisition in Transformers
di: Billa, Jayadev
Pubblicazione: (2026)

Decomposing the Depth Profile of Fine-Tuning
di: Billa, Jayadev
Pubblicazione: (2026)

When Audio-LLMs Don't Listen: A Cross-Linguistic Study of Modality Arbitration
di: Billa, Jayadev
Pubblicazione: (2026)

The Cascade Equivalence Hypothesis: When Do Speech LLMs Behave Like ASR$\rightarrow$LLM Pipelines?
di: Billa, Jayadev
Pubblicazione: (2026)

Beyond Steering Vector: Flow-based Activation Steering for Inference-Time Intervention
di: Jin, Zehao, et al.
Pubblicazione: (2026)

VSPO: Vector-Steered Policy Optimization for Behavioral Control
di: Zhang, Xuechen, et al.
Pubblicazione: (2026)

Beyond Multiple Choice: Evaluating Steering Vectors for Summarization
di: Braun, Joschka, et al.
Pubblicazione: (2025)

White-Box Sensitivity Auditing with Steering Vectors
di: Cyberey, Hannah, et al.
Pubblicazione: (2026)

Tell Your Model Where to Attend: Post-hoc Attention Steering for LLMs
di: Zhang, Qingru, et al.
Pubblicazione: (2023)

Personalized Steering of Large Language Models: Versatile Steering Vectors Through Bi-directional Preference Optimization
di: Cao, Yuanpu, et al.
Pubblicazione: (2024)

Steer2Adapt: Dynamically Composing Steering Vectors Elicits Efficient Adaptation of LLMs
di: Han, Pengrui, et al.
Pubblicazione: (2026)

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

When and Why Does Unsupervised RL Succeed in Mathematical Reasoning? A Manifold Envelopment Perspective
di: Zhang, Zelin, et al.
Pubblicazione: (2026)

Shifting Perspectives: Steering Vectors for Robust Bias Mitigation in LLMs
di: Siddique, Zara, et al.
Pubblicazione: (2025)

No Training Wheels: Steering Vectors for Bias Correction at Inference Time
di: Gupta, Aviral, et al.
Pubblicazione: (2025)

Understanding How CodeLLMs (Mis)Predict Types with Activation Steering
di: Lucchetti, Francesca, et al.
Pubblicazione: (2024)

SteerConf: Steering LLMs for Confidence Elicitation
di: Zhou, Ziang, et al.
Pubblicazione: (2025)

Scaling Up Membership Inference: When and How Attacks Succeed on Large Language Models
di: Puerto, Haritz, et al.
Pubblicazione: (2024)

Understanding Unreliability of Steering Vectors in Language Models: Geometric Predictors and the Limits of Linear Approximations
di: Braun, Joschka
Pubblicazione: (2026)

In-context Vectors: Making In Context Learning More Effective and Controllable Through Latent Space Steering
di: Liu, Sheng, et al.
Pubblicazione: (2023)

Conceptors for Semantic Steering
di: Triantafyllopoulos, Ilias, et al.
Pubblicazione: (2026)

Textual Steering Vectors Can Improve Visual Understanding in Multimodal Large Language Models
di: Gan, Woody Haosheng, et al.
Pubblicazione: (2025)

Behavioral Steering in a 35B MoE Language Model via SAE-Decoded Probe Vectors: One Agency Axis, Not Five Traits
di: Yap, Jia Qing
Pubblicazione: (2026)

Towards Understanding Steering Strength
di: Taimeskhanov, Magamed, et al.
Pubblicazione: (2026)

Where is the signal in tokenization space?
di: Geh, Renato Lui, et al.
Pubblicazione: (2024)

MorphNAS: Differentiable Architecture Search for Morphologically-Aware Multilingual NER
di: Devadiga, Prathamesh, et al.
Pubblicazione: (2025)

Fractional Reasoning via Latent Steering Vectors Improves Inference Time Compute
di: Liu, Sheng, et al.
Pubblicazione: (2025)

Steering Language Models with Weight Arithmetic
di: Fierro, Constanza, et al.
Pubblicazione: (2025)

Steering Language Models With Activation Engineering
di: Turner, Alexander Matt, et al.
Pubblicazione: (2023)

HyperSteer: Activation Steering at Scale with Hypernetworks
di: Sun, Jiuding, et al.
Pubblicazione: (2025)

A Unified Understanding and Evaluation of Steering Methods
di: Im, Shawn, et al.
Pubblicazione: (2025)

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

Compositional Steering of Large Language Models with Steering Tokens
di: Radevski, Gorjan, et al.
Pubblicazione: (2026)

ROAST: Rollout-based On-distribution Activation Steering Technique
di: Su, Xuanbo, et al.
Pubblicazione: (2026)

Differentially Private Steering for Large Language Model Alignment
di: Goel, Anmol, et al.
Pubblicazione: (2025)

Compute Where it Counts: Self Optimizing Language Models
di: Akhauri, Yash, et al.
Pubblicazione: (2026)

Spherical Steering: Geometry-Aware Activation Rotation for Language Models
di: You, Zejia, et al.
Pubblicazione: (2026)

Guiding Giants: Lightweight Controllers for Weighted Activation Steering in LLMs
di: Hegazy, Amr, et al.
Pubblicazione: (2025)

Analyze Feature Flow to Enhance Interpretation and Steering in Language Models
di: Laptev, Daniil, et al.
Pubblicazione: (2025)