---
title: "AI & Machine Learning"
canonical: https://papersays.com/hubs/ai-ml
date: 2026-09-30
hub: "AI & Machine Learning"
---

# AI & Machine Learning

How modern AI systems are built, evaluated and made efficient, without the hype.

## Papers

- [Speculative Decoding Under Real Serving Load](https://papersays.com/p/speculative-decoding-under-real-serving-load): Speculative decoding promises faster text generation. This preprint measures what happens once real traffic, batching and memory limits enter the picture.
- [Teaching Agents to Explore When Rewards Are Rare](https://papersays.com/p/teaching-agents-to-explore-when-rewards-are-rare): A simulation study tests whether a simple curiosity bonus can help reinforcement learning agents find rewards that appear only once in a long while.
- [Do Long Context Windows Actually Get Used?](https://papersays.com/p/do-long-context-windows-actually-get-used): Models advertise windows of a million tokens. A new benchmark asks how much of that window they can really use when it counts.