works on

From the 2 of 7 linked papers with an AI index.

activity
20242026
collaborators

6 papers

cs.AI2026

Attractor Geometry of Transformer Memory: From Conflict Arbitration to Confident Hallucination

Qiyao Liang, Risto Miikkulainen, Ila Fiete

Language models draw on two knowledge sources: facts baked into weights (parametric memory, PM) and information in context (working memory, WM). We study two mechanistically distin…

cs.LG2026

Evolution Strategies at Scale: LLM Fine-Tuning Beyond Reinforcement Learning

Xin Qiu, Yulu Gan, Conor F. Hayes +6

The paper shows that evolution strategies can successfully fine‑tune billion‑parameter large language models without backpropagation, outperforming reinforcement learning in stabil…

cs.LG2026

The Blessing of Dimensionality in LLM Fine-tuning: A Variance-Curvature Perspective

Qiyao Liang, Jinyeop Song, Yizhou Liu +4

Weight-perturbation evolution strategies (ES) can fine-tune billion-parameter language models with surprisingly small populations (e.g., ), contradicting classical…

physics.bio-ph2025

Modular connectivity in neural networks emerges from Poisson noise-motivated regularisation, and promotes robustness and compositional generalisation

Daoyuan Qian, Qiyao Liang, Ila Fiete

Circuits in the brain commonly exhibit modular architectures that factorise complex tasks, resulting in the ability to compositionally generalise and reduce catastrophic forgetting…

cs.LG2025

Compositional Generalization via Forced Rendering of Disentangled Latents

Qiyao Liang, Daoyuan Qian, Liu Ziyin +1

Composition-the ability to generate myriad variations from finite means-is believed to underlie powerful generalization. However, compositional generalization remains a key challen…

cs.AI2024

How Diffusion Models Learn to Factorize and Compose

Qiyao Liang, Ziming Liu, Mitchell Ostrow +1

Diffusion models are capable of generating photo-realistic images that combine elements which likely do not appear together in the training set, demonstrating the ability to \texti…