@lateinteraction: Intuition: Compaction is agentic recurrence (RNNs), whereas recursion (RLMs) is agentic attention. Recurrence maintains…
Summary
The article shares an intuition that compaction in RNNs represents agentic recurrence while recursion in RLMs represents agentic attention, comparing different context-handling approaches in AI models.
View Cached Full Text
Cached at: 08/15/26, 07:50 PM
Intuition: Compaction is agentic recurrence (RNNs), whereas recursion (RLMs) is agentic attention.
Recurrence maintains a constant-size state at all times by compressing the past.
Attention keeps all context fully represented and then decides how to re-process it for each step.
Similar Articles
@TDataScience: Follow along @neural_avb's all-in-one deep dive to learn "what recursive language models (RLMs) are, why they are winni…
An educational deep dive into recursive language models (RLMs), explaining what they are, why they are winning long-context benchmarks, and how they differ from existing agentic harness designs like ReAct or CodeAct, using a simple case study.
Recursive Language Models
This paper introduces Recursive Language Models (RLMs), an inference strategy that enables LLMs to process arbitrarily long prompts by treating them as external environments and recursively calling themselves over prompt snippets. RLMs handle inputs two orders of magnitude beyond context windows and outperform base LLMs on long-context tasks with comparable cost.
alexzhang13/rlm
Recursive Language Models (RLMs) introduce a task-agnostic inference paradigm enabling language models to handle near-infinite contexts by recursively calling themselves over input, with an accompanying open-source inference engine and training environment.
RecMem: Recurrence-based Memory Consolidation for Efficient and Effective Long-Running LLM Agents
RecMem is a recurrence-based memory consolidation method for long-running LLM agents that reduces token consumption by up to 87% while improving accuracy, by only invoking LLMs when semantically similar interactions recur.
@cwolferesearch: Agentic RL requires new algorithm modifications. In GRPO, the “group” used starts to change when training agents… In va…
This thread discusses modifications to GRPO for agentic RL, focusing on different levels of advantage normalization (prompt-level, task-level, environment-level) to handle higher reward variance in multi-task, multi-turn environments.