For employers

Hire this AI Trainer

Sign in or create an account to invite AI Trainers to your job.

LoginCreate account
Invite to Job
Travis Lu

Travis Lu

Professor of Mathematics

USA flagLos Angeles, Usa
$50.00/hrExpertInternal Proprietary Tooling

Key Skills

Software

Internal/Proprietary Tooling

Top Subject Matter

No subject matter listed

Top Data Types

Computer Code ProgrammingComputer Code Programming
ImageImage
TextText

Top Task Types

Data Collection
Prompt Response Writing SFT
Question Answering
RLHF
Text Generation

Freelancer Overview

I am a mathematical logic and foundations expert with 25 years of academic experience as a Professor of Mathematics at Peking University. Since 2020, I have specialized in AI reasoning evaluation, creating high-quality datasets for RLHF and supervised fine-tuning. My work includes annotating and validating AI-generated proofs, logical puzzles, and foundational paradoxes, ensuring coherence, rigor, and correctness. I bring a unique combination of theoretical expertise and practical annotation experience that supports high-standard AI training and evaluation.

ExpertEnglishChinese Mandarin

Labeling Experience

SFT Dataset Creation for Logical Puzzles and Paradoxes

Internal Proprietary ToolingTextQuestion AnsweringText Generation
Authored a specialized Supervised Fine-Tuning (SFT) dataset focused on foundational mathematical logic, complex puzzles, and philosophical paradoxes. The goal was to train models to handle abstract and non-standard reasoning. My work involved crafting high-quality prompt-response pairs, including detailed explanations and meta-commentary on common logical pitfalls and problem-solving strategies.

Authored a specialized Supervised Fine-Tuning (SFT) dataset focused on foundational mathematical logic, complex puzzles, and philosophical paradoxes. The goal was to train models to handle abstract and non-standard reasoning. My work involved crafting high-quality prompt-response pairs, including detailed explanations and meta-commentary on common logical pitfalls and problem-solving strategies.

2021

AI Mathematical Reasoning Evaluation & RLHF

Internal Proprietary ToolingTextRelationshipQuestion Answering
`Evaluated the logical rigor and validity of mathematical proofs generated by advanced large language models. My responsibilities included: Ranking and comparing multiple AI-generated solutions based on correctness, clarity, and axiomatic consistency (RLHF). Authoring "gold-standard" proofs and detailed, step-by-step solutions for use in supervised fine-tuning (SFT) datasets. Developing and applying detailed rubrics to score the logical coherence and identify subtle fallacies in AI reasoning. Providing meta-analytical commentary on error patterns to guide model development.`

`Evaluated the logical rigor and validity of mathematical proofs generated by advanced large language models. My responsibilities included: Ranking and comparing multiple AI-generated solutions based on correctness, clarity, and axiomatic consistency (RLHF). Authoring "gold-standard" proofs and detailed, step-by-step solutions for use in supervised fine-tuning (SFT) datasets. Developing and applying detailed rubrics to score the logical coherence and identify subtle fallacies in AI reasoning. Providing meta-analytical commentary on error patterns to guide model development.`

2020

Education

M

MIT

Bachelor of Science, Mathematics (Honors)

Bachelor of Science
Not specified
U

University of Chicago

Master of Science, Mathematics (Logic & Foundations)

Master of Science
Not specified

Work History

N

N/A

Mathematical Logic & Reasoning Consultant

Remote
2020 - Present
P

Peking University

Professor of Mathematics

Beijing
1995 - 2020