EP247: PEEK Cures AI Goldfish Memory

EP247: PEEK Cures AI Goldfish Memory

Title: PEEK: Context Map as an Orientation Cache for Long-Context LLM Agents Source: http://arxiv.org/abs/2605.19932v1 Summary: This work introduces 'context maps' as a novel architectural primitive for long-context agents, enabling them to cache and maintain structured orientation knowledge about recurring external datasets. By implementing a programmable cache policy for distilling and translating inference-time signals, it significantly improves efficiency and accuracy across multi-turn reasoning workloads.

EP246: Replacing AI manuals with programmable runtimes

Title: Formal Skill: Programmable Runtime Skills for Efficient and Accurate LLM Agents Source: http://arxiv.org/abs/2605.19604v1 Summary: This work introduces a foundational architectural primitive for agents that replaces informal natural-language instructions with programmable, stateful runtime skills governed by hook policies and action schemas. This shift from prompting to executable state machines provides a more enforceable and token-efficient control surface for reliable agentic workflows in real-world environments.

EP245: The Geometric Shape of AI Reasoning

Title: A Measure-Theoretic Analysis of Reasoning: Structural Generalization and Approximation Limits Source: http://arxiv.org/abs/2605.19944v1 Summary: This paper establishes fundamental theoretical bounds for LLM reasoning, proving that scaling physical layer depth is a non-negotiable requirement for out-of-distribution generalization that cannot be bypassed by scaling width. It also formalizes why specific architectural choices, such as shift-invariant embeddings, are mathematically necessary to maintain reasoning equivariance across domain shifts.

EP244: Training decentralized AI through private handoffs

Title: Learning to Hand Off: Provably Convergent Workflow Learning under Interface Constraints Source: http://arxiv.org/abs/2605.19140v1Summary: This research provides the first finite-sample guarantee for neural Q-learning in decentralized multi-agent settings, a foundational breakthrough for reliable agentic workflow learning. By formalizing handoffs as interface-constrained SMDPs, it enables provably convergent learning in complex LLM pipelines where agents have restricted observability.

EP243: Breaking the AI data wall with SYNPRO

Title: Generating Pretraining Tokens from Organic Data for Data-Bound ScalingSource: http://arxiv.org/abs/2605.17849v1 Summary: This work addresses the transition of LLM pretraining into data-bound regimes by introducing a synthetic data generation framework that maximizes the utility of limited organic datasets. It represents a significant breakthrough in scaling laws, demonstrating how to unlock up to 5x more effective tokens through model-aware rephrasing and reformatting.