Yiliya

Key Skills

Software

Other

Label Studio

Appen

Top Subject Matter

LLM output evaluation and bilingual text labeling (EN/ZH)

AI-assisted content QA and instruction-following evaluation

Text quality assurance for AI-generated content

Top Data Types

Text

Image

Top Task Types

RLHF

Evaluation/Rating

Red Teaming

Prompt + Response Writing (SFT)

Classification

Entity (NER) Classification

Freelancer Overview

Freelance AI Output Reviewer & Text Annotator. Brings 2+ years of professional experience across complex text processing, long-form content editing, and quality-focused execution. Core strengths include RLHF Preference Ranking , Instruction-Following Review , and Adversarial Dialogue Testing . Education includes Bachelor of Arts, Jiliang University (Expected 2028 / In Progress) and Russian-Language Preparatory Program, Belarusian State University (2025). AI-training focus includes data types such as Text and labeling workflows including Evaluation, Rating, and RLHF.

Labeling Experience

Freelance AI Output Reviewer & Text Annotator

Don't discloseTextEvaluation/Rating

Freelance AI output reviewer and text annotator for Chinese crowdsourcing platforms. Rated chatbot responses for quality and instruction-following, and labeled short texts by topic and sentiment while writing concise rationale notes for each decision. Tested model reasoning via logically connected follow-up questions and documented inconsistencies, while also reviewing Chinese-English bilingual outputs for translation accuracy and natural phrasing. • Ensure responses matched user instructions including edge cases • Score accuracy, helpfulness, tone, and factual consistency • Identify logical contradictions and unsupported claims • Maintain consistent judgment across large batches and edge cases

2025 - Present

Author & Creative Director — AI-Assisted Novel Project

OtherTextRLHF

Directed and iteratively reviewed a 150,000-character AI-assisted science fiction novel project with a focus on evaluating AI-generated drafts. Assessed logical consistency, narrative coherence, and instruction-following accuracy chapter by chapter, using structured outlines, specifications, and revision passes to correct model issues. Applied observations about how language models produce plausible but logically flawed content to improve annotation-style judgment. • Evaluate AI draft outputs for internal logic and coherence • Verify adherence to story rules and detailed specifications • Rewrite and re-edit AI-generated chapters for correctness • Translate QA-style checks into preference ranking and output validation thinking

2024 - Present

Freelance Content Editor

Don't discloseTextText Generation

Freelance content editor using AI-generated text as the primary input. Edited roughly 60,000 words with a focus on catching logical inconsistencies, unclear phrasing, and factual-sounding but unsupported claims. Produced structured notes describing each change using a rationale format aligned with annotation QA practices. • Detect contradictions across paragraphs and hedged statements • Identify unsupported or overconfident factual claims • Improve clarity, tone, and natural phrasing • Document edits in structured, explanation-focused notes

2023 - Present

Education

Work History