Entity · technique

representation-level steering

techniqueactiverepresentation-level-steering-ecb97e81·1 events·first seen May 29, 2026

Aliases: representation-level steering

Co-occurring entities

reinforcement learning with belief-state rewards Contextual Belief Management (CBM)BeliefTrack Zhejiang University NLP Group (ZJUNLP)

More like this (12)

representation-level sensitive information leakage Activation Steering SafeSteer State-Conditioned Dynamic Steering On-Policy Distillation for LLM Safety: A Routing Approach to Template-Robust Realignment Agentic Chain-of-Thought Steering Selective Proxy Distillation Steerable Model Merging Visual Verification Enables Inference-time Steering and Autonomous Policy Improvement Agentic Chain-of-Thought Steering for Efficient and Controllable LLM Reasoning steering vectors Forecasting With LLMs: Improved Generalization Through Feature Steering

Recent events (1)

6arXiv · cs.CL·May 29, 2026·source ↗

BeliefTrack: Benchmarking and Improving Contextual Belief Management in LLMs

This paper introduces Contextual Belief Management (CBM) as a framework for studying how LLMs should update, preserve, or ignore information across long-horizon interactions. The authors release BeliefTrack, a closed-world benchmark with symbolic verifiers enabling exact turn-level evaluation across Rule Discovery and Circuit Diagnosis tasks. Vanilla LLMs show severe CBM failures; reinforcement learning with belief-state rewards reduces failure rates by 70.9% on average, while representation-level steering achieves 46.1% reduction. Probing experiments reveal latent belief-state dynamics underlying these failures.

Evaluation and Benchmarking Agent and Tool Ecosystem reinforcement learning with belief-state rewards Contextual Belief Management (CBM)BeliefTrack +3 more