LLM Digest

Story

arxiv_cs_lg · Sep 3, 2026 · paper

Source brief

RATL: Learning from Retrieved Residuals for Robust Multivariate Time-Series Forecasting

arxiv.orgSep 3, 2026
original source linked

In brief

Retrieval-augmented generation (RAG) complements parametric models with retrieved external evidence. The same idea is attractive for continuous-output regression, but directly reusing retrieved target values is often...

Feed lens

eval

Read the original at arxiv.org →Open in live feed Read that day’s brief

RATL: Learning from Retrieved Residuals for Robust Multivariate Time-Series Forecasting

Earlier in this thread 4 items

When Does Online Adaptation Pay on the Edge? A Leakage-Free Evaluation of Warmup, Learning-Rate Selection, and Resource Trade-offs for Time-Series Forecasting

SimSkill: A Lifelong Learning AI Agent for Autonomous Mastery of Traffic Simulation

Online Reinforcement Learning in the Met Office Unified Model through Distributed Model-Agent Coupling

SCoNE: Selective Context-aware Neuron Editing for Robust Retrieval-Augmented Generation