Ax Zhibin Wang, Zetao Hong, Xue Li, Zibo Wang, Shipeng Li, Qingkai Meng, Qing Wang, Chengying Huan, Rong Gu, Sheng Zhong, Chen Tian 5/5/2026

STAR: Decode-Phase Rescheduling for LLM Inference

STAR dynamic rescheduling algorithm for LLM decode-phase workload balancing addressing SLO violations and OOM failures from variable output lengths.

Ax Weihao Bo, Shan Zhang, Yanpeng Sun, Jingjing Wu, Qunyi Xie, Xiao Tan, Kunbin Chen, Wei He, Xiaofan Li, Na Zhao, Jingdong Wang, Zechao Li 5/5/2026

Agentic Learner with Grow-and-Refine Multimodal Semantic Memory

Agentic learner system with grow-and-refine multimodal semantic memory addressing MLLM limitations by retaining domain knowledge across multimodal problem-solving.

Ax Jingkang Wang, Henry Che, Yun Chen, Ze Yang, Lily Goli, Sivabalan Manivasagam, Raquel Urtasun 5/5/2026

Flux4D: Flow-based Unsupervised 4D Reconstruction

Flux4D uses flow-based unsupervised learning for 4D dynamic scene reconstruction without motion annotations, addressing scalability of NeRF and Gaussian Splatting.

Ax Dongseok Kim, Hyoungsun Choi, Mohamed Jismy Aashik Rasool, Gisung Oh 5/5/2026

$\phi$-Table: A Statistical Explanation for Global SHAP

Phi-table statistical explanation method extending SHAP for tabular model interpretability with directional summaries and uncertainty quantification.

Ax Hongjie Wang, Niraj K. Jha 5/5/2026

LinMU: Multimodal Understanding Made Linear

LinMU proposes a Vision-Language Model with linear-complexity attention mechanism for efficient edge deployment and high-resolution image/video processing.

Ax Mame Diarra Toure, David A. Stephens 5/5/2026

Singular Bayesian Neural Networks

Bayesian neural networks with singular posteriors via low-rank weight parameterization reducing parameters from O(mn) to O(r(m+n)).

Ax Varun Ursekar (Emily), Apaar Shanker (Emily), Veronica Chatrath (Emily), Yuan (Emily), Xue, Sam Denton 5/5/2026

VeRO: An Evaluation Harness for Agents to Optimize Agents

VeRO evaluation harness for assessing coding agent performance on agent optimization through iterative edit-execute-evaluate cycles.

HN kilian-ai 5/5/2026

An LLM agent that runs on any Linux box

Portable LLM agent framework running in shell script with streaming chat, tool calling, memory, and mentor mode. No dependencies beyond curl and jq.

HN goranmoomin 5/5/2026

Amp's GPT 5.5 Model Analysis

Product analysis of GPT-5.5's capabilities as coding agent with improved tool use and constraint handling.