AIResearch

DR Tulu: Evolving Rubrics Teach AI to Do Deep Research

Kari Jaaskelainen

25 Nov 2025 — 1 min read

Long, well-sourced answers are hard for AI to learn: most training rewards short, easily graded Q&A. This work introduces Reinforcement Learning with Evolving Rubrics (RLER) - grading guides that co-evolve with the model, so feedback stays aligned with what the model actually explores during multi-step research.

Built with RLER, Deep Research Tulu (DR Tulu-8B) is the first open model directly trained for open-ended, long-form deep research.

Produces multi-step, cited, long-form answers with on-policy feedback.
Outperforms open deep research models on four benchmarks (science, healthcare, general domains).
Matches or exceeds proprietary systems while being smaller and cheaper per query.
Everything is open: data, models, code, plus a new MCP-based agent infrastructure.

Read the paper: https://arxiv.org/abs/2511.19399v1

Paper: https://arxiv.org/abs/2511.19399v1

Register: https://www.AiFeta.com

AIResearch ReinforcementLearning LLMs OpenSource DeepResearch NLP Science Healthcare

Read more

Beyond URLs: Metadata That Makes LLMs Train Faster

Beyond URLs: Metadata That Makes LLMs Train Faster

Smarter LLMs, Faster—thanks to metadata What if training a large language model didn't just rely on text, but on the context around it? This study shows that adding fine-grained metadata—not just URLs—can meaningfully speed up pretraining and improve quality. * Beyond URLs: detailed quality signals (e.

Escaping the Verifier: Learning to Reason via Demonstrations

Escaping the Verifier: Learning to Reason via Demonstrations

LLMs can learn to reason—without task verifiers Many real-world problems don’t have automatic checkers to grade answers, even though we have lots of expert solutions. RARO (Relativistic Adversarial Reasoning Optimization) shows how to train reasoning skills from those examples alone. How it works: * A policy (the model) tries

VacuumVLA: A Two-in-One Robot Hand That Grips and Suctions

VacuumVLA: A Two-in-One Robot Hand That Grips and Suctions

Robots guided by Vision-Language-Action (VLA) AI are getting better at everyday tasks—but most still use simple two-finger grippers. That limits them on smooth, flat, or handleless objects. VacuumVLA is a low-cost robot hand that merges a standard two-finger gripper with a vacuum suction cup. The robot can switch between

ChatDRex: No‑Code, Conversational AI for Drug Repurposing

ChatDRex: No‑Code, Conversational AI for Drug Repurposing

Meet ChatDRex Finding new uses for approved drugs can save years and millions. But making those predictions usually takes teams of specialists and a tangle of tools. ChatDRex changes that. It’s a conversation-based, no‑code, multi‑agent system that lets clinicians and researchers ask complex bioinformatics questions in plain