Copertina

The Logical Implication Steering Method for Conditional Interventions on Transformer Generation

Fuente: arXiv

Salvato in:

Dettagli Bibliografici
Autore principale:	Kalajdzievski, Damjan
Natura:	Preprint
Pubblicazione:	2025
Soggetti:	Machine Learning
Accesso online:	Acceder al recurso
Tags:	Aggiungi Tag Nessun Tag, puoi essere il primo ad aggiungerne!!

Documenti analoghi

Scaling Laws for Forgetting When Fine-Tuning Large Language Models
di: Kalajdzievski, Damjan
Pubblicazione: (2024)

Towards Steering without Sacrifice: Principled Training of Steering Vectors for Prompt-only Interventions
di: Bao, Yuntai, et al.
Pubblicazione: (2026)

Amortized Network Intervention to Steer the Excitatory Point Processes
di: Song, Zitao, et al.
Pubblicazione: (2023)

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

Temporal Logic Specification-Conditioned Decision Transformer for Offline Safe Reinforcement Learning
di: Guo, Zijian, et al.
Pubblicazione: (2024)

Programming Refusal with Conditional Activation Steering
di: Lee, Bruce W., et al.
Pubblicazione: (2024)

Conditional Logical Message Passing Transformer for Complex Query Answering
di: Zhang, Chongzhi, et al.
Pubblicazione: (2024)

Analyzing the Generalization and Reliability of Steering Vectors
di: Tan, Daniel, et al.
Pubblicazione: (2024)

MidSteer: Optimal Affine Framework for Steering Generative Models
di: Gaintseva, Tatiana, et al.
Pubblicazione: (2026)

Multi-Attribute Steering of Language Models via Targeted Intervention
di: Nguyen, Duy, et al.
Pubblicazione: (2025)

Tractable Transformers for Flexible Conditional Generation
di: Liu, Anji, et al.
Pubblicazione: (2025)

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

Feynman-Kac-Flow: Inference Steering of Conditional Flow Matching to an Energy-Tilted Posterior
di: Mark, Konstantin, et al.
Pubblicazione: (2025)

Faithful Bi-Directional Model Steering via Distribution Matching and Distributed Interchange Interventions
di: Bao, Yuntai, et al.
Pubblicazione: (2026)

Graph Diffusion Transformers for Multi-Conditional Molecular Generation
di: Liu, Gang, et al.
Pubblicazione: (2024)

Relational Rank Geometry in Transformers: Detecting and Steering Hidden-State Relation Frames
di: Kobrosly, Mazen
Pubblicazione: (2026)

Intervention and Conditioning in Causal Bayesian Networks
di: Galhotra, Sainyam, et al.
Pubblicazione: (2024)

General and Efficient Steering of Unconditional Diffusion
di: Wang, Qingsong, et al.
Pubblicazione: (2026)

Circuit Transformer: A Transformer That Preserves Logical Equivalence
di: Li, Xihan, et al.
Pubblicazione: (2024)

Steering Without Breaking: Mechanistically Informed Interventions for Discrete Diffusion Language Models
di: Zhou, Hanhan, et al.
Pubblicazione: (2026)

SafeSteer: Interpretable Safety Steering with Refusal-Evasion in LLMs
di: Ghosh, Shaona, et al.
Pubblicazione: (2025)

Improving Chain-of-Thought for Logical Reasoning via Attention-Aware Intervention
di: Phuong, Nguyen Minh, et al.
Pubblicazione: (2026)

FishBack: Pullback Fisher Geometry for Optimal Activation Steering in Transformers
di: Wang, Sihan, et al.
Pubblicazione: (2026)

Generative and Nonparametric Approaches for Conditional Distribution Estimation: Methods, Perspectives, and Comparative Evaluations
di: Chin, Yen-Shiu, et al.
Pubblicazione: (2026)

Conditional Variable Flow Matching: Transforming Conditional Densities with Amortized Conditional Optimal Transport
di: Generale, Adam P., et al.
Pubblicazione: (2024)

Steering Safely or Off a Cliff? Rethinking Specificity and Robustness in Inference-Time Interventions
di: Goyal, Navita, et al.
Pubblicazione: (2026)

Steered Generation via Gradient-Based Optimization on Sparse Query Features
di: Bhattacharyya, Sumanta, et al.
Pubblicazione: (2026)

Lagrangian Perturbation Diffusion Steering: Latent Reinforcement Learning for Generative Policies
di: Simsir, Hikmet, et al.
Pubblicazione: (2026)

Where to Steer: Input-Dependent Layer Selection for Steering Improves LLM Alignment
di: Gadgil, Soham, et al.
Pubblicazione: (2026)

A Logic for Expressing Log-Precision Transformers
di: Merrill, William, et al.
Pubblicazione: (2022)

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

Steering Generative Models with Experimental Data for Protein Fitness Optimization
di: Yang, Jason, et al.
Pubblicazione: (2025)

A Framework for Low-Effort Training Data Generation for Urban Semantic Segmentation
di: Zavadski, Denis, et al.
Pubblicazione: (2025)

Exploring Diverse Generation Paths via Inference-time Stiefel Activation Steering
di: Zhu, Dongxuan, et al.
Pubblicazione: (2026)

Molecules Meet Language: Confound-Aware Representation Learning and Chemical Property Steering in Transformer-VAE Latent Spaces
di: Elabid, Zakaria, et al.
Pubblicazione: (2026)

Generalization on the Unseen, Logic Reasoning and Degree Curriculum
di: Abbe, Emmanuel, et al.
Pubblicazione: (2023)

Manifold-Guided Attention Steering
di: Li, Ian, et al.
Pubblicazione: (2026)

Spectral Conditioning of Attention Improves Transformer Performance
di: Saratchandran, Hemanth, et al.
Pubblicazione: (2026)

Minimal-Intervention KV Retention via Set-Conditioned Diversity
di: Sun, Libo, et al.
Pubblicazione: (2026)

Steer-to-Detect: Probing Hidden Representations for Detection of LLM-Generated Texts
di: Liang, Luxu, et al.
Pubblicazione: (2026)