Ax Jingpu Cheng, Ping Liu, Qianxiao Li, Chi Zhang 3/30/2026

Machine Unlearning under Retain-Forget Entanglement

Machine unlearning framework addressing retain-forget entanglement where retained samples unintentionally affected by forgetting correlated features.

Ax Roope Niemi, Anastasiia Petrovych, Arghya Ranjan Das, Enrico Lupi, Chang Sun, Dimitrios Danopoulos, Marlon Joshua Helbing, Mia Liu, Sebastian Dittmeier, Michael Kagan, Vladimir Loncar, Maurizio Pierini 3/30/2026

PQuantML: A Tool for End-to-End Hardware-aware Model Compression

PQuantML: open-source hardware-aware neural network compression library for pruning and quantization with unified interface for latency-constrained deployment.

Ax Jun Yang, Yuechun Sun, Yi Wu, Rodrigo Caridad, Yongwei Yuan, Jianan Yao, Shan Lu, Kexin Pei 3/30/2026

ExVerus: Verus Proof Repair via Counterexample Reasoning

LLM framework for formal proof repair using counterexample-guided reasoning and behavioral feedback to improve automated verification.

Ax Afonso Simpl\'icio, Gon\c{c}alo Vinagre, Miguel Moura Ramos, Diogo Tavares, Rafael Ferreira, Giuseppe Attanasio, Duarte M. Alves, In\^es Calvo, In\^es Vieira, Rui Guerra, James Furtado, Beatriz Canaverde, Iago Paulo, Vasco Ramos, Diogo Gl\'oria-Silva, Miguel Faria, Marcos Treviso, Daniel Gomes, Pedro Gomes, David Semedo, Andr\'e Martins, Jo\~ao Magalh\~aes 3/30/2026

AMALIA Technical Report: A Fully Open Source Large Language Model for European Portuguese

AMALIA: fully open source LLM trained on high-quality European Portuguese data with native evaluation benchmark and improved pt-PT representation.

Ax Dung V. Nguyen, Hieu M. Vu, Nhi Y. Pham, Lei Zhang, Tan M. Nguyen 3/30/2026

Activation Steering with a Feedback Controller

Control-theoretic framework for LLM activation steering with feedback controllers, connecting empirical steering methods to proportional control theory for safety alignment.

Ax Tiansheng Wen, Yifei Wang, Aosong Feng, Long Ma, Xinyang Liu, Yifan Wang, Lixuan Guo, Bo Chen, Stefanie Jegelka, Chenyu You 3/30/2026

Route Experts by Sequence, not by Token

Sequence-level TopK (SeqTopK) improves Mixture-of-Experts routing in LLMs by adapting expert assignment per sequence rather than per token without retraining.

Ax R Sri Prakash, Nikhil Karamchandani, Sharayu Moharir 3/30/2026

Cascading Bandits With Feedback

Cascading Bandits analyzes decision-making policies for edge inference with multiple models, providing theoretical regret guarantees for Explore-then-Commit and Thompson Sampling approaches.

Ax Yassir Bendou, Omar Ezzahir, Eduardo Fernandes Montesuma, Gabriel Mahuas, Victoria Shevchenko, Mike Gartrell 3/30/2026

ReBaPL: Repulsive Bayesian Prompt Learning

Repulsive Bayesian Prompt Learning addresses overfitting in prompt learning for foundation models using Bayesian inference framework for improved out-of-distribution generalization.

Ax Zhenchao Tang, Fang Wang, Haohuai He, Jiale Zhou, Tianxu Lv, Jun Zhu, Shouzhi Chen, Minghao Yang, Yu Wang, Jiayang Wu, Yidong Song, Yaokun Li, Jiehui Huang, Dawei Huang, Zhi Song, Jianhua Yao 3/30/2026

Aligning LLMs with Biomedical Knowledge using Balanced Fine-Tuning

Balanced Fine-Tuning aligns LLMs with biomedical knowledge through confidence-weighted token-level optimization and adaptive reward mechanisms.