Sicong Jiang

LLM-as-Judge for Open-ended Tasks: Researching on rubric-based LLM evaluators for open-ended outputs and trajectories, and systematically probing judge failure modes to support trustworthy agent self-improvement.

Self-evolving Agent: Building agents that iteratively improve their policies via self-refinement loops, with a focus on reliable feedback signals and long-horizon behavior.

Research: As a founding member of the Research team, I lead benchmarking and evaluation for agentic and multimodal LLMs. I led the EditReward project and co-developed large-scale benchmarks including SuperGPQA and VeriWeb.

Advanced Dataset & Pipeline Design: Led several zero-to-one pipeline builds—architecting and deploying high-difficulty dataset solutions and production pipelines from scratch across coding, IMO-level math, multimodal data, agentic trajectories, and RL environments. These datasets and pipelines are directly used for model training and evaluation for multiple frontier AI labs.

Multimodal Data Pipelines: Built data pipelines and multi-stage QA systems for multimodal LLM projects, overseeing large-scale annotation workflows and label consistency.

Dataset Quality & Validation: Conducted analysis and validation to refine annotations and ensure robust datasets for LLM post-training.

AgentThink (Agent Reasoning): Led a collaboration with Xiaomi and Tsinghua on tool-augmented reasoning for vision-language models in autonomous driving, achieving +54% answer accuracy on open-source models.

Adversarial LLM4TS: Developed a black-box attack framework and public benchmarks for LLM-based time-series forecasting, in collaboration with the Amazon Chronos and Nixtla teams.

Multi-Agent RL Exploration: Developed a multi-agent search strategy combining MADDPG with frontier-based exploration, and built evaluation benchmarks for exploration efficiency.