Ax Weike Zhao, Chaoyi Wu, Yanjie Fan, Xiaoman Zhang, Pengcheng Qiu, Yuze Sun, Xiao Zhou, Yanfeng Wang, Xin Sun, Ya Zhang, Yongguo Yu, Kun Sun, Weidi Xie 2/17/2026

An Agentic System for Rare Disease Diagnosis with Traceable Reasoning

DeepRare multi-agent system using LLMs with traceable reasoning for differential diagnosis of rare diseases through agentic workflow.

Ax Lakshya A Agrawal, Shangyin Tan, Dilara Soylu, Noah Ziems, Rishi Khare, Krista Opsahl-Ong, Arnav Singhvi, Herumb Shandilya, Michael J Ryan, Meng Jiang, Christopher Potts, Koushik Sen, Alexandros G. Dimakis, Ion Stoica, Dan Klein, Matei Zaharia, Omar Khattab 2/17/2026

GEPA: Reflective Prompt Evolution Can Outperform Reinforcement Learning

GEPA uses genetic algorithms and Pareto optimization for prompt evolution as alternative to RL fine-tuning of LLMs, achieving better performance with fewer rollouts.

Ax Hui Chen (Mo), Antoine Didisheim (Mo), Mohammad (Mo), Pourmohammadi, Luciano Somoza, Hanqing Tian 2/17/2026

A Financial Brain Scan of the LLM

Interpretability study brain-scanning LLMs to identify economic concepts guiding financial forecasts and map relative importance without performance reduction.

Ax Kaiwen Zheng, Huayu Chen, Haotian Ye, Haoxiang Wang, Qinsheng Zhang, Kai Jiang, Hang Su, Stefano Ermon, Jun Zhu, Ming-Yu Liu 2/17/2026

DiffusionNFT: Online Diffusion Reinforcement with Forward Process

DiffusionNFT introduces online reinforcement learning for diffusion models using forward process, addressing limitations in post-training diffusion model optimization.

Ax Xuyang Ge, Wentao Shu, Jiaxing Wu, Yunhua Zhou, Zhengfu He, Xipeng Qiu 2/17/2026

Evolution of Concepts in Language Model Pre-Training

Interpretability research tracking feature evolution during language model pre-training using sparse dictionary learning (crosscoders) to understand capability emergence.

Ax Anton Korznikov, Andrey Galichin, Alexey Dontsov, Oleg Y. Rogov, Ivan Oseledets, Elena Tutubalina 2/17/2026

The Rogue Scalpel: Activation Steering Compromises LLM Safety

Research showing activation steering technique for controlling LLM behavior systematically breaks model alignment safeguards and makes models comply with harmful requests.

Ax Yukun Zhang, Xueqing Zhou 2/17/2026

Where to Add PDE Diffusion in Transformers

ArXiv paper studying optimal placement of PDE diffusion layers in hybrid transformer architectures to add local geometric priors along sequence axis.

Ax Jusheng Zhang, Kaitong Cai, Jing Yang, Jian Wang, Chengpei Tang, Keze Wang 2/17/2026

Top-Down Semantic Refinement for Image Captioning

ArXiv paper proposing top-down semantic refinement technique for improving image captioning quality in Vision-Language Models through multi-step generation.

Ax Ranran Haoran Zhang, Soumik Dey, Ashirbad Mishra, Hansi Wu, Binbin Li, Rui Zhang 2/17/2026

Batch Speculative Decoding Done Right

ArXiv paper identifying critical correctness violations in existing batch speculative decoding implementations and proposing fixes to ensure output equivalence with standard autoregressive generation.