Jingxuan Fan

4 papers · 2025–2026 · 4 venues · across top CS/AI venues

Authors are matched by name, so researchers who share a name ("Yang Liu", "Wei Wang") appear as a single profile and their paper counts, venues and achievements are combined. How identity is handled.

Achievements

Jump to papers ↓

🌍 Conference Polyglot · Bronze 🌉 Interdisciplinary Bridge · Bronze 🗺️ Taxonomy Completionist (11)

Venues

AAAI (1) CVPR (1) ICLR (1) ICML (1)

Top co-authors

Mingye Gao (2) Xiaomin Li (2) Xu Pan (1) Michael Brenner (1) Eric Hanchen Jiang (1) Binxu Wang (1) Sarah Martinson (1) Kaylie Hausknecht (1) Nianli Peng (1) Danxian Liu (1)

Research topics

Artificial Intelligence (3) Machine Learning (3) Core AI (3) Optimization & Theory (3) Computer Vision (1) Deep Learning (1) Natural Language Processing (1) Applications (1) Core Methods (1) Generation (1) Models (1) Techniques (1)

Keywords

mathematical reasoning (1) text-to-image generation (1) safety alignment (1) mutual information (1) bradley-terry model (1) benchmark dataset (1) reward hacking (1) reward model (1) mechanistic interpretability (1) chain-of-thought prompting (1) text encoder (1) cross attention (1) diffusion transformer (1) spatial relation (1) large language model (1) entropy-guided reward (1) multi-head reward model (1) applied mathematics (1) analytical approximation (1) rlhf (1)

Papers

Circuit Mechanisms for Spatial Relation Generation in Diffusion Transformers CVPR 2026 ENCORE: Entropy-guided Reward Composition for Multi-head Safety Reward Models AAAI 2026 HARDMath: A Benchmark Dataset for Challenging Problems in Applied Mathematics ICLR 2025 RuleAdapter: Dynamic Rules for training Safety Reward Models in RLHF ICML 2025