publications

2026

1. ACL Oral

   Why Are We Moral? An LLM-based Agent Simulation Approach to Study Moral Evolution

   Zhou Ziheng, Huacong Tang, Mingjie Bi, and 7 more authors

   Proceedings of the Annual Meeting of the Association for Computational Linguistics (ACL), 2026

   🔥 Oral Abs arXiv Website

   true

   LLM-based agent simulation framework modeling prehistoric hunter-gatherer societies to investigate moral evolution. Accepted as ACL 2026 main conference oral.

2. Preprint

   Less is More: Early Stopping Rollout for On-Policy Distillation

   Zhou Ziheng, Jiaqi Li, Huacong Tang, and 2 more authors

   arXiv preprint arXiv:2605.27028, 2026

   Abs arXiv

   Identifies an Off-policy Teacher Decay problem in on-policy distillation, where teacher models lose effectiveness on later tokens once the student’s earlier trajectory diverges from training data. Proposes Early Stopping Rollout (ESR), which restricts generation to initial response tokens, outperforming full-rollout approaches across model sizes and tasks while improving computational efficiency. Analyzes underlying mechanisms including Cascading Alignment and Sub-mode Commitment.

3. ICLR WS Spotlight

   SciCrafter: Can Current Language Models Close the Discovery-to-Application Loop?

   Zhou Ziheng, Huacong Tang, Jinyuan Zhang, and 9 more authors

   ICLR 2026 Workshop RSI / Under Review for COLM 2026, 2026

   🔥 Spotlight Abs Website

   true

   We introduce SciCrafter, a Minecraft-based task suite that operationalizes the discovery-to-application loop through parameterized redstone circuit tasks. Evaluating frontier models including GPT-5.2, Gemini-3-Pro, and Claude-Opus-4.5, we find that all plateau at approximately 25% success rate.

4. ICLR WS

   Credibility Governance: A Social Mechanism for Collective Self-Correction under Weak Truth Signals

   Wanying He, Yanxi Lin, Zhou Ziheng, and 5 more authors

   ICLR 2026 Workshop AIMS, 2026

   Abs arXiv

   Introduces Credibility Governance (CG), a mechanism that reallocates influence by identifying agents and viewpoints that consistently align with evolving evidence, outperforming vote-based and stake-weighted approaches across misinformation, noise, and adversarial pressure scenarios.

5. ACL Findings

   Simple Role Assignment is Extraordinarily Effective for Safety Alignment

   Zhou Ziheng, Jiakun Ding, Zhaowei Zhang, and 6 more authors

   Findings of the Association for Computational Linguistics (ACL Findings), 2026

   Abs arXiv Website

   Proposes role conditioning grounded in Theory of Mind as a compact alternative to principle-based alignment. Reduces unsafe outputs on the WildJailbreak benchmark from 81.4% to 3.6% with DeepSeek-V3.

6. Under Review

   Why Current AI Architectures Cannot Support Consciousness: Mediated Self-Access and the Need for Ontological Identity

   Ruosen Gao and Zhou Ziheng

   Under Review, INQUIRE 2026, 2026
7. ICLR

   On the Generalization of SFT: A Reinforcement Learning Perspective with Reward Rectification

   Yongliang Wu, Yizhou Zhou, Zhou Ziheng, and 7 more authors

   International Conference on Learning Representations (ICLR), 2026

   Abs arXiv

   Proposes Dynamic Fine-Tuning (DFT) to stabilize gradient updates for SFT. A single-line code change significantly outperforms standard SFT. Accepted as ICLR 2026 Poster.

2025

1. Under Review

   A Human-centric Framework for Debating the Ethics of AI Consciousness Under Uncertainty

   Zhou Ziheng, Haiqiang Dai, Bin Ling, and 2 more authors

   arXiv preprint arXiv:2512.02544, 2025

   Abs arXiv

   Proposes a three-level ethical framework for machine consciousness questions, establishing five factual determinations about AI consciousness.

2. ICLR WS

   Learning to Be Fair: Modeling Fairness Dynamics by Simulating Moral-Based Multi-Agent Resource Allocation

   Haiyan Feng, Yuqiao Du, Huacong Tang, and 5 more authors

   ICLR 2026 Workshop AFAA, 2025

   Abs

   Investigates fairness as an evolving, negotiated construct using LLM-driven cognitive agents in multi-agent resource allocation simulations.

3. Preprint

   A Hybrid Approach to Video World Modeling: Bridging Pixel and Latent Spaces

   Zhou Ziheng, Taichang Zhou, Xianhao Wang, and 8 more authors

   2025

   Abs

   Proposes a hybrid video world model that bridges pixel-space and latent-space generation via object-centric representations, enabling efficient action planning in embodied AI.

4. Preprint

   Who Is Us: A Generative Theory of Moral Psychology from Group Boundary Dynamics

   Zhou Ziheng

   Working paper, 2025

   Abs

   A generative theory of moral psychology grounded in group boundary dynamics.

5. Preprint

   What is Intelligence

   Zhou Ziheng

   Working paper, 2025

   Abs

   A theoretical exploration of the nature of intelligence.

2023

1. arXiv

   Aligner: One Global Token is Worth Millions of Parameters When Aligning Large Language Models

   Zhou Ziheng, Yingnian Wu, Song-Chun Zhu, and 1 more author

   arXiv preprint arXiv:2312.05503, 2023

   Abs arXiv

   A parameter-efficient fine-tuning method where a single globally shared token with merely 5,000 parameters performs comparably to LoRA which requires millions of parameters.

2. ICCV

   ARNOLD: A Benchmark for Language-Grounded Task Learning With Continuous States in Realistic 3D Scenes

   Ran Gong, Jiangyong Huang, Yizhou Zhao, and 9 more authors

   Proceedings of the IEEE/CVF International Conference on Computer Vision (ICCV), 2023

   arXiv
3. Nat. Med.

   Early Detection of Visual Impairment in Young Children Using a Smartphone-based Deep Learning System

   Wenben Chen, Ruiyang Li, Qinji Yu, and 9 more authors

   Nature Medicine, 2023
4. ISBI

   MDT-Net: Multi-Domain Transfer by Perceptual Supervision for Unpaired Images in OCT Scan

   Weinan Song, Gaurav Fotedar, Nima Tajbakhsh, and 3 more authors

   2023 IEEE 20th International Symposium on Biomedical Imaging (ISBI), 2023

2020

1. BMJ

   Artificial Intelligence-Enabled Screening for Diabetic Retinopathy: A Real-World, Multicenter and Prospective Study

   Yifan Zhang, Jianfeng Shi, Yingfeng Peng, and 8 more authors

   BMJ Open Diabetes Research & Care, 2020

2019

1. Nat. BME

   Discrimination of the Behavioural Dynamics of Visually Impaired Infants Via Deep Learning

   Erping Long, Zhenzhen Liu, Yi Xiang, and 9 more authors

   Nature Biomedical Engineering, 2019