Targeted Retrieval, Compact Representations: How CoT Reasoning Improves Long-Context Counting
cs.AI, cs.CL, cs.LG, stat.AP
Submitted: 2026-09-30
Updated: 2026-09-30
Code: https://github.com/Twist-Shan/count-state-geometry-suppl
Terminology
Sources
- Belief Dynamics Reveal the Dual Nature of In-Context Learning and Activation Steering
- Learning Dynamics of Chain-of-Thought State Tracking in a Solvable Transformer Model
- The Right Answer, the Wrong Direction: Why Transformers Fail at Counting and How to Fix It
- Gemma 4 Technical Report
- How to use and interpret activation patching
- Language Models Use Trigonometry to Do Addition
- Scaling Laws for Neural Language Models
- Symmetry in language statistics shapes the geometry of model representations
- Ministral 3
- The Hydra Effect: Emergent Self-repair in Language Model Computations
- NVIDIA Nemotron 3: Efficient and Open Intelligence
- NVIDIA Nemotron Nano 2: An Accurate and Efficient Hybrid Mamba-Transformer Reasoning Model
- MemGPT: Towards LLMs as Operating Systems
- Nemotron Elastic: Towards Efficient Many-in-One Reasoning LLMs
- ChatGLM: A Family of Large Language Models from GLM-130B to GLM-4 All Tools
- Steering Language Models With Activation Engineering
- How Does Unfaithful Reasoning Emerge from Autoregressive Training? A Study of Synthetic Experiments
- Task Vector Geometry Underlies Dual Modes of Task Inference in Transformers
- Qwen3 Technical Report
Related papers
- MAVEN-T: Reinforced Heterogeneous Distillation for Real-Time Multi-Agent Trajectory Prediction
- Model Discovery Agent: LLM-assisted Bayesian experiment design for data-efficient discovery of mechanistic world models
- The Clinician's Veto: Navigating Trust, Liability, and Uncertainty in Autonomous AI Prescribing
- MindHelper: Closed-Loop Embodied Mental-State Reasoning for Precision Intervention
- Incumbent Advantage: Brand Bias and Cognitive Manipulation Dynamics in LLM Recommendation Systems
- VSAL: A Vision Solver with Adaptive Layouts for Graph Property Detection