Hi, I'm

Demi

王若晗

Researching LLMs, Agents & Reasoning at CMU LTI.
Building toward the agentic future.

Scroll

Recent

2026-06-01Starting my internship at Microsoft Research in Redmond!
2026-05-26Presenting 1 paper at the Agent Skills Workshop @ CAIS 2026 in San Jose!
2025-08-25Excited to start my master's journey at Carnegie Mellon University!
2025-07-28We won 2nd place at Kimi track in AdventureX 2025!
2025-02-11UGround was accepted as ICLR 2025 Oral (1.8%)!
2025-01-06I'm on China's top 1 intellectual game show, The Brain Season 12! Check it out on YouTube🙌

Research

CAIS 2026 · Agent Skills Workshop

Skill Induction for Code Agents on Web Automation

Exploring code-native skill induction on web agents, where skills take the form of standalone Playwright functions. Building a multi-agent pipeline that decouples solving, verification, and updating — reaching 67.2% on WebArena-Verified, a 6.4 pp improvement over the no-skill code agent baseline and 10.3 pp above programmatic skill + action-based agent.

Web AgentsCode AgentsSkill InductionWeb Automation
Paper
MSRA · 2025 · Under Submission

Understanding RL Updates for LLM Reasoning

Exploring efficient and effective reinforcement learning for LLM reasoning with verifiable rewards. We analyze what drives RLVR updates under different off-policy settings and propose Adaptive Clip Policy Optimization (ACPO), an adaptive clipping method that improves reasoning performance across benchmarks and model scales.

RLVRGRPOLLMReasoning
PaperCode
ByteDance · 2024

Multimodal LLM-based Violation Detection

Improving e-commerce content moderation with multimodal large language models for violation detection in real-world systems. Production deployment at scale, achieving 30%+ precision improvement and reducing manual review workload by 40%.

Multimodal LLMIn-Context LearningContent ModerationIndustry

Building

2026

Cagen

Agent operating system for organizations and enterprise growth. Early-stage AI startup backed by MiraclePlus and CMU AI Venture Studio.

Website
Startup
2026

Wisp

Natural language workflow automation engine. Explores once, compiles into deterministic workflows, and re-runs in seconds. Built at YC Browser Use Web Agent Hackathon.

DemoGitHubTwitter
Hackathon
2025

Miko

AI-native Desktop Pet for Productivity. 2nd place winner at the AdventureX 2025 Hackathon Kimi Track.

DemoGitHub
Hackathon
2023

LifeBuddy

Personal lifestyle assistant for budget-friendly living. 2nd place winner at the Baidu & Founder Park AGI Hackathon.

Hackathon

Researcher.
Builder.
Explorer.

Life is a giant playground — I'm here to unlock every ride:

SnowboardClimbDanceRaveSolo TripFashion
Hip-HopWritingCocktailsPokerPuzzles