Cover Image

Steering LLM Reasoning Through Bias-Only Adaptation

Fuente: arXiv

Saved in:

Bibliographic Details
Main Authors:	Sinii, Viacheslav, Gorbatovski, Alexey, Cherepanov, Artem, Shaposhnikov, Boris, Balagansky, Nikita, Gavrilov, Daniil
Format:	Preprint
Published:	2025
Subjects:	Machine Learning Artificial Intelligence
Online Access:	Acceder al recurso
Tags:	Add Tag No Tags, Be the first to tag this record!

Similar Items

F-GRPO: Don't Let Your Policy Learn the Obvious and Forget the Rare
by: Plyusov, Daniil, et al.
Published: (2026)

Trust-Region Behavior Blending for On-Policy Distillation
by: Plyusov, Daniil, et al.
Published: (2026)

Small Vectors, Big Effects: A Mechanistic Study of RL-Induced Reasoning via Steering Vectors
by: Sinii, Viacheslav, et al.
Published: (2025)

The Differences Between Direct Alignment Algorithms are a Blur
by: Gorbatovski, Alexey, et al.
Published: (2025)

Learn Your Reference Model for Real Good Alignment
by: Gorbatovski, Alexey, et al.
Published: (2024)

You Do Not Fully Utilize Transformer's Representation Capacity
by: Gerasimov, Gleb, et al.
Published: (2025)

ESSA: Evolutionary Strategies for Scalable Alignment
by: Korotyshova, Daria, et al.
Published: (2025)

Linear Transformers with Learnable Kernel Functions are Better In-Context Models
by: Aksenov, Yaroslav, et al.
Published: (2024)

Next Embedding Prediction Makes World Models Stronger
by: Bredis, George, et al.
Published: (2026)

Enhancing Vision-Language Model Training with Reinforcement Learning in Synthetic Worlds for Real-World Success
by: Bredis, George, et al.
Published: (2025)

Teach Old SAEs New Domain Tricks with Boosting
by: Koriagin, Nikita, et al.
Published: (2025)

Train One Sparse Autoencoder Across Multiple Sparsity Budgets to Preserve Interpretability and Accuracy
by: Balagansky, Nikita, et al.
Published: (2025)

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

Mechanistic Permutability: Match Features Across Layers
by: Balagansky, Nikita, et al.
Published: (2024)

Unraveling the Complexity of Memory in RL Agents: an Approach for Classification and Evaluation
by: Cherepanov, Egor, et al.
Published: (2024)

A New Perspective on Transformers in Online Reinforcement Learning for Continuous Control
by: Kachaev, Nikita, et al.
Published: (2025)

Re:Frame -- Retrieving Experience From Associative Memory
by: Zelezetsky, Daniil, et al.
Published: (2025)

Guided Star-Shaped Masked Diffusion
by: Meshchaninov, Viacheslav, et al.
Published: (2025)

Kronecker Factorization Improves Efficiency and Interpretability of Sparse Autoencoders
by: Kurochkin, Vadim, et al.
Published: (2025)

In-Context Reinforcement Learning for Variable Action Spaces
by: Sinii, Viacheslav, et al.
Published: (2023)

Diffusion Language Models Generation Can Be Halted Early
by: Vaina, Sofia Maria Lo Cicero, et al.
Published: (2023)

Memory, Benchmark & Robots: A Benchmark for Solving Complex Tasks with Reinforcement Learning
by: Cherepanov, Egor, et al.
Published: (2025)

KAGE-Bench: Fast Known-Axis Visual Generalization Evaluation for Reinforcement Learning
by: Cherepanov, Egor, et al.
Published: (2026)

Recurrent Action Transformer with Memory
by: Cherepanov, Egor, et al.
Published: (2023)

Improving GFlowNets with Monte Carlo Tree Search
by: Morozov, Nikita, et al.
Published: (2024)

ELMUR: External Layer Memory with Update/Rewrite for Long-Horizon RL Problems
by: Cherepanov, Egor, et al.
Published: (2025)

Memory Retention Is Not Enough to Master Memory Tasks in Reinforcement Learning
by: Shchendrigin, Oleg, et al.
Published: (2026)

The Rogue Scalpel: Activation Steering Compromises LLM Safety
by: Korznikov, Anton, et al.
Published: (2025)

Don't Blind Your VLA: Aligning Visual Representations for OOD Generalization
by: Kachaev, Nikita, et al.
Published: (2025)

BarrierSteer: LLM Safety via Learning Barrier Steering
by: Tran, Thanh Q., et al.
Published: (2026)

OGLS-SD: On-Policy Self-Distillation with Outcome-Guided Logit Steering for LLM Reasoning
by: Yang, Yuxiao, et al.
Published: (2026)

Learning Shortest Paths with Generative Flow Networks
by: Morozov, Nikita, et al.
Published: (2026)

Act Only When It Pays: Efficient Reinforcement Learning for LLM Reasoning via Selective Rollouts
by: Zheng, Haizhong, et al.
Published: (2025)

Steered LLM Activations are Non-Surjective
by: Mishra, Aayush, et al.
Published: (2026)

Directional Attractors in LLM Reasoning: How Similarity Retrieval Steers Iterative Summarization Based Reasoning
by: Tekin, Cagatay, et al.
Published: (2025)

Test-time Diverse Reasoning by Riemannian Activation Steering
by: Khanh, Ly Tran Ho, et al.
Published: (2025)

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

PeerRank: Autonomous LLM Evaluation Through Web-Grounded, Bias-Controlled Peer Review
by: Margalit, Yanki, et al.
Published: (2026)

Graph-Regularized Sparse Autoencoders for LLM Safety Steering
by: Yeon, Jehyeok, et al.
Published: (2025)

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