Trained Agentic Context Management

cs.LG, cs.CL

Submitted: 2026-10-01

Updated: 2026-10-01

Comments: 17 pages, 6 figures, 4 tables. Code: https://github.com/brycesandlund/infinite-context. Under review

Code: https://github.com/brycesandlund/infinite-context

License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/

Sources

• Recursive Language Models Meet Uncertainty: The Surprising Effectiveness of Self-Reflective Program Search for Long Context
• Titans: Learning to Memorize at Test Time
• Oolong: Evaluating Long Context Reasoning and Aggregation Capabilities
• Language Models are Few-Shot Learners
• Transformers are SSMs: Generalized Models and Efficient Algorithms Through Structured State Space Duality
• DeepSeek-V3.2: Pushing the Frontier of Open Large Language Models
• Recursive Agent Optimization
• Gemma 2: Improving Open Language Models at a Practical Size
• Accelerating scientific discovery with Co-Scientist
• The Llama 3 Herd of Models
• DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via Reinforcement Learning
• Transformers are RNNs: Fast Autoregressive Transformers with Linear Attention
• Kimi K2.5: Visual Agentic Intelligence
• Mamba-3: Improved Sequence Modeling using State Space Principles
• MiniMax Sparse Attention
• SnapKV: LLM Knows What You are Looking for Before Generation
• The Complexity Trap: Simple Observation Masking Is as Efficient as LLM Summarization for Agent Context Management
• Understanding R1-Zero-Like Training: A Critical Perspective
• Recursive Agent Harnesses

Related papers

• Polynomial-Augmented Neural Networks (PANNs) with Weak Orthogonality Constraints for Enhanced Function and PDE Approximation
• AIRL-S: Unifying Reinforcement Learning and Search-Based Test-Time Scaling via Adversarial Inverse Reinforcement Learning
• Transformers as Bayesian In-Context Experimenters: Smoothness-Adaptive Efficient ATE Estimation
• Convergence issues in Relational Concept Analysis based on AOC-posets
• Beliefs Beyond Posteriors: The Predictive Cost of ELBO Binding
• Understanding Diffusion Models via Ratio-Based Function Approximation with SignReLU Networks