0

Faeze Brahman

Research scientist at Allen AI; works on instruction tuning, safety, and evaluation; co-author on Tulu 3 and many AI2 post-training papers.

Role
researcher
Papers
19

Cite

Notes

Only stored in your browser.

19papers

Authored papers

19

Meta-Reinforcement Learning with Self-Reflection for Agentic Search

arXiv 2026

2026

Olmo 3

arXiv 2025

2025

IssueBench: Millions of Realistic Prompts for Measuring Issue Bias in LLM Writing Assistance

arXiv 2025

2025

DR Tulu: Reinforcement Learning with Evolving Rubrics for Deep Research

arXiv 2025

2025

Aligning LLMs to Ask Good Questions A Case Study in Clinical Reasoning

arXiv 2025

2025

Large-Scale Data Selection for Instruction Tuning

arXiv 2025

2025

Tulu 3: Pushing Frontiers in Open Language Model Post-Training

preprint

2024

2 OLMo 2 Furious

arXiv 2024

2024

WildTeaming at Scale: From In-the-Wild Jailbreaks to (Adversarially) Safer Language Models

arXiv 2024

2024

Hybrid Preferences: Learning to Route Instances for Human vs. AI Feedback

arXiv 2024

2024

RESTOR: Knowledge Recovery through Machine Unlearning

arXiv 2024

2024

WildBench: Benchmarking LLMs with Challenging Tasks from Real Users in the Wild

arXiv 2024

2024

Agent Lumos: Unified and Modular Training for Open-Source Language Agents

arXiv 2023

2023

Leftover Lunch: Advantage-based Offline Reinforcement Learning for Language Models

arXiv 2023

2023

Inference-Time Policy Adapters (IPA): Tailoring Extreme-Scale LMs without Fine-tuning

arXiv 2023

2023

PlaSma: Making Small Language Models Better Procedural Knowledge Models for (Counterfactual) Planning

arXiv 2023

2023

STEER: Unified Style Transfer with Expert Reinforcement

arXiv 2023

2023

In Search of the Long-Tail: Systematic Generation of Long-Tail Inferential Knowledge via Logical Rule Guided Search

arXiv 2023

2023

ParsiNLU: A Suite of Language Understanding Challenges for Persian

arXiv 2020

2020

Affiliations

Currently at

Allen Institute for AI (Ai2)

researcher · non profit

Frequent co-authors

10

from 19 papers