# Freshness and the Limits of Heuristic Trend Detection in Temporal RAG

arXiv: https://arxiv.org/abs/2509.19376 · HTML: https://mlsanity.com/paper/2509.19376/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Builds on 1 of its 18 references that are in the map.
- First posted on arXiv 20 Sep 2025.

## Builds on (most cited first, 1 of 1)

- [Reading Between the Timelines: RAG for Answering Diachronic Questions](https://mlsanity.com/paper/2507.22917/) (2025)

## Built on by (most cited first, 1 of 1)

- [Reliable Post-Retrieval Assembly for Agent Memory: Separating Evidence Extraction from Policy Execution](https://mlsanity.com/paper/2606.01435/) (2026)

## Abstract (arXiv)

We present a lightweight, model-agnostic temporal layer for RAG and use cybersecurity data to separate two problems that are usually conflated. For freshness, a half-life recency prior surfaces the newest relevant item where a cosine-only baseline scores 0.00; on a hard NVD CVE test, where the freshest item is not the most similar, it reaches Latest@10 of 0.60 versus 0.20 for a semantic-then-newest baseline, but stays partial and parameter-sensitive. For topic evolution, a heuristic tracker's low 0.08 macro-F1 is driven by the labeling rule, not the clusterer (HDBSCAN: 0.10; fixing the rule alone reaches 0.49, and 0.96 without clustering noise). We contribute a reproducible decoupling of the two, with honest real-data scope and a reference implementation.