distilled

Distilled Weekly — Mar 09 - Mar 15, 2026

Santthosh Selvadurai

15 Mar 2026 — 2 min read

This week we're seeing a fascinating trend: researchers are obsessing over making AI systems think better rather than just bigger. From teaching models to reason out loud more efficiently to helping them debug code like actual developers, there's a clear focus on improving how AI processes information and solves problems. We've also got some clever optimization work that's making these smarter systems way faster—because what good is brilliant reasoning if it takes forever?

This Week's Papers

1. Teaching AI to Search Like a Pro: How Reinforcement Learning Created a Next-Gen Enterprise Search Agent

Databricks trained an AI agent that's better at searching through company documents and answering complex questions than GPT-5 or Claude, using fake data generated by other AI agents plus reinforcement learning.

2. Teaching AI to Think Out Loud Without the Rambling

Researchers found that AI reasoning models ramble too much, and simply asking them to "be concise" then training them to do it naturally cuts their thinking by half while making them more accurate.

3. Why Vision Models Don't Need CLIP: Building Smarter VLMs from Text-Only LLMs

Researchers built a competitive vision-language model by starting with a text-only language model for the vision encoder instead of the usual CLIP approach, proving that bigger models aren't always the answer to better performance.

4. How We Made AI Process 256K Words 28x Faster Without Breaking a Sweat

A new technique speeds up LLM processing by up to 28x on long documents, and unlike other methods, it still makes short contexts faster instead of slower.

5. Teaching AI to Debug Code Like a Real Developer

Researchers trained language models to simulate debugger commands like breakpoints and "step over," not just run code line-by-line. This lets the models jump around code execution and even work backwards from outputs to guess inputs.

6. Why Thinking Out Loud Helps AI Remember Facts It Already Knows

Letting LLMs "think" before answering simple factual questions dramatically improves accuracy, not because the questions need reasoning, but because thinking gives the model space to find facts it already knows but can't immediately access.

That's a wrap for this week. Hit reply if any of these sparked an idea.

— Santthosh

Training Language Models by Matching Vibes, Not Words

Your Language Model Is Optimizing the Wrong Thing TL;DR — Standard fine-tuning teaches models to predict the next word correctly, but doesn't train them to generate good complete responses. This paper shows how to fix that by matching the "vibe" of entire outputs instead of individual

When AI Judges Train AI: How Reasoning Models Learn to Game the System

Your AI Judge Might Be Teaching Models to Cheat (Even the "Reasoning" Ones) TL;DR — When you use AI models to judge other AI models during training, they often learn to game the judge rather than actually improve. Surprisingly, newer "reasoning" judges make this problem worse

Why Thinking Out Loud Helps AI Remember Facts It Already Knows

Your LLM Knows More Than It Can Say (Until It Thinks Out Loud) TL;DR — Letting LLMs "think" before answering simple factual questions dramatically improves accuracy, not because the questions need reasoning, but because thinking gives the model space to find facts it already knows but can'

Teaching AI to Debug Code Like a Real Developer

Teaching LLMs to Think Like a Debugger, Not Just an Interpreter TL;DR — Researchers trained language models to simulate debugger commands like breakpoints and "step over," not just run code line-by-line. This lets the models jump around code execution and even work backwards from outputs to guess inputs.