Recursive Language Models
PulseAugur coverage of Recursive Language Models — every cluster mentioning Recursive Language Models across labs, papers, and developer communities, ranked by signal.
-
RLM-Rust cuts LLM token costs by 96% with recursive models
A developer has created RLM-Rust, a tool designed to significantly reduce the token costs associated with processing large context windows in AI applications. By implementing Recursive Language Models (RLM) in Rust, the…
-
New research explores modular and recursive methods for automatic prompt optimization
Two new research papers introduce novel methods for optimizing prompts used with large language models. The first, SAPO, breaks down prompts into segments like role, context, and task, allowing for targeted improvements…
-
Meta launches Muse Code; Google DeepMind reshuffles leadership; Anthropic builds chip team
Several major AI players are undergoing significant structural changes and strategic expansions. Meta has launched Muse Code, a terminal coding agent powered by Muse Spark 1.2, capable of complex repository-level engine…
-
TimeRLM uses recursive language models for precise time-series anomaly detection
Researchers have developed TimeRLM, a novel recursive language model designed to improve anomaly localization in long-context time-series data. This approach addresses the performance degradation seen in traditional tim…
-
Recursion's role in long-context language models debated in new paper
A new paper explores the effectiveness of recursion in Recursive Language Models (RLMs) for handling long contexts. Researchers tested SRLM, which samples eight context-interaction programs and selects the most confiden…
-
DSPy's new RLM module tackles LLM context rot
The DSPy framework has introduced a new module, dspy.RLM, designed to address the issue of "context rot" in large language models. This phenomenon, identified by MIT researchers, describes how models lose performance on…
-
Ypi: A Recursive Coding Agent Built on Raspberry Pi
Ypi is a new recursive coding agent that utilizes Raspberry Pi and is inspired by Recursive Language Models. Named after the Y Combinator, Ypi is designed to be a self-referential Pi agent.
-
Apple unveils SRLM to enhance long-context language model reasoning
Apple Machine Learning Research has introduced a new framework called Self-Reflective Program Search for Long Context (SRLM). This framework aims to improve how language models handle long contexts by using uncertainty-…
-
Proposal uses semantic compression for AI long-context sessions
A proposal suggests using semantic compression as an input diffusion technique to handle AI sessions longer than the current context window. This method treats the context like a progressive render, starting with a comp…
-
New context compression method uses diffusion noise function
An independent researcher is proposing a novel method for handling extremely long contexts in language models by treating context compression as a diffusion noise function. This approach involves multiple passes over th…
-
Amazon Bedrock AgentCore breaks context window limits for LLMs
Amazon Bedrock has introduced AgentCore, a new capability designed to overcome the limitations of context windows in large language models. This feature enables models to process and reason over documents of virtually a…
-
RAG system architectures show varied robustness to knowledge base poisoning
Researchers have investigated the vulnerability of Retrieval-Augmented Generation (RAG) systems to knowledge base poisoning, finding that system architecture significantly impacts adversarial robustness. Evaluations on …