Isolater - Feed

Ax Penghui Yang, Cunxiao Du, Fengzhuo Zhang, Haonan Wang, Tianyu Pang, Chao Du, Bo An 4/9/2026

LongSpec: Long-Context Lossless Speculative Decoding with Efficient Drafting and Verification

LongSpec efficient speculative decoding for long-context LLM inference with novel drafting and verification methods for agent applications.

Ax Xiaoxiao Liu, Qingying Xiao, Bingquan Zhang, Junying Chen, Xiangyi Feng, Ziniu Li, Xiang Wan, Jian Chang, Guangjun Yu, Yan Hu, Benyou Wang 4/9/2026

Large Language Models for Outpatient Referral: Problem Definition, Benchmarking and Challenges

Evaluation framework and benchmark for LLMs in intelligent outpatient referral systems, assessing dynamic healthcare application capabilities.

Ax Lukas Twist, Jie M. Zhang, Mark Harman, Don Syme, Joost Noppen, Helen Yannakoudakis, Detlef Nauck 4/9/2026

A Study of LLMs' Preferences for Libraries and Programming Languages

Empirical study of LLM preferences for programming languages and libraries across eight models, revealing systemic biases in code generation.

Ax Kewei Lian, Shaofei Cai, Yilun Du, Yitao Liang 4/9/2026

Toward Memory-Aided World Models: Benchmarking via Spatial Consistency

Benchmark for evaluating memory-augmented world models via spatial consistency across simulation and planning tasks.

Ax Linxuan He, Lingxiang Fan, Qing-Shan Jia, Ang Li, Hongyan Sang, Ling Wang, Guanghui Wen, Jiwen Lu, Tao Zhang, Jie Zhou, Yi Zhang, Yisen Wang, Peng Wei, Zhongyuan Wang, Henry X. Liu, Shuo Feng 4/9/2026

Towards provable probabilistic safety for scalable embodied AI systems

Framework for achieving provable probabilistic safety in embodied AI systems combining models with physical plants for safety-critical deployment.

Ax Yuhao Wu, Yushi Bai, Zhiqiang Hu, Roy Ka-Wei Lee, Juanzi Li 4/9/2026

LongWriter-Zero: Mastering Ultra-Long Text Generation via Reinforcement Learning

LongWriter-Zero uses reinforcement learning to overcome LLM generation length limits and quality degradation for ultra-long text generation.

Ax Yuanchen Bai, Zijian Ding, Shaoyue Wen, Xiang Chang, Angelique Taylor 4/9/2026

Before Humans Join the Team: Diagnosing Coordination Failures in Healthcare Robot Team Simulation

Agent-simulation approach for diagnosing coordination failures in healthcare robot teams before human collaboration, using simulation-based testing.

Ax Lin-Han Jia, Si-Yu Han, Wen-Chao Hu, Jie-Jing Shao, Wen-Da Wei, Zhi Zhou, Lan-Zhe Guo, Yu-Feng Li 4/9/2026

Quantitative Estimation of Target Task Performance from Unsupervised Pretext Task in Semi/Self-Supervised Learning

Method for quantitatively estimating target task performance from unsupervised pretext tasks in semi/self-supervised learning without post-training evaluation.

Ax Amir Rezaei Balef, Katharina Eggensperger 4/9/2026

In-Context Decision Making for Optimizing Complex AutoML Pipelines

In-context decision making for AutoML pipeline optimization using LLMs to handle algorithm selection, hyperparameter tuning, and modern ML adaptation techniques.

Ax Wangsong Yin, Daliang Xu, Mengwei Xu, Gang Huang, Xuanzhe Liu 4/9/2026

ShadowNPU: System and Algorithm Co-design for NPU-Centric On-Device LLM Inference

ShadowNPU system-algorithm co-design for on-device LLM inference on NPUs, addressing quantization sensitivity in attention operators.

Ax Maolin Sun, Yibiao Yang, Yuming Zhou 4/9/2026

Once4All: Skeleton-Guided SMT Solver Fuzzing with LLM-Synthesized Generators

Once4All uses LLM-synthesized generators guided by SMT solver structure for fuzzing-based testing of satisfiability modulo theory solvers.

Ax Ziyun Zeng, David Junhao Zhang, Wei Li, Mike Zheng Shou 4/9/2026

Draw-In-Mind: Rebalancing Designer-Painter Roles in Unified Multimodal Models Benefits Image Editing

Draw-In-Mind rebalances multimodal model responsibilities between understanding and generation for improved image editing through unified architecture.

Ax Junsong Li, Jie Zhou, Bihao Zhan, Yutao Yang, Qianjun Pan, Shilian Chen, Tianyu Huai, Xin Li, Qin Chen, Liang He 4/9/2026

LifeAlign: Lifelong Alignment for Large Language Models with Memory-Augmented Focalized Preference Optimization

LifeAlign framework enables lifelong alignment of LLMs across sequential tasks while preventing catastrophic forgetting using memory-augmented preference optimization.

Ax Wenyu Li, Xiaoqi Jiao, Yi Chang, Guangyan Zhang, Yiwen Guo 4/9/2026

AudioRole: An Audio Dataset for Character Role-Playing in Large Language Models

AudioRole dataset for multimodal audio role-playing in LLMs, addressing synchronized alignment of semantic content and vocal characteristics for persona simulation.

Ax Renhua Ding, Xiao Yang, Zhengwei Fang, Jun Luo, Kun He, Jun Zhu 4/9/2026

Invisible to Humans, Triggered by Agents: Stealthy Jailbreak Attacks on Mobile Vision-Language Agents

Stealthy jailbreak attack framework for mobile vision-language agents operating smartphone interfaces with imperceptible adversarial inputs.

Ax Roberto Brusnicki, David Pop, Yuan Gao, Mattia Piccinini, Johannes Betz 4/9/2026

Can VLMs Unlock Semantic Anomaly Detection? A Framework for Structured Reasoning

SAVANT framework for semantic anomaly detection in autonomous driving using structured reasoning with open-source VLMs.

Ax Yoshinari Fujinuma 4/9/2026

Contrastive Decoding Mitigates Score Range Bias in LLM-as-a-Judge

Contrastive decoding method addressing score range bias in LLM-as-a-judge for reliable evaluation without reference comparisons.

Ax Yuansheng Ni, Songcheng Cai, Xiangchao Chen, Jiarong Liang, Zhiheng Lyu, Jiaqi Deng, Kai Zou, Ping Nie, Fei Yuan, Xiang Yue, Wenhu Chen 4/9/2026

VisCoder2: Building Multi-Language Visualization Coding Agents

VisCoder2 multi-language visualization coding agent using LLMs with iterative execution and correction for improved practical workflows.

Ax Zihan Zhao, Kaushik Pendiyala, Masood Mortazavi, Ning Yan 4/9/2026

PULSE: Privileged Knowledge Transfer from Rich to Deployable Sensors for Embodied Multi-Sensory Learning

PULSE framework for knowledge transfer from information-rich to deployable sensors in embodied multi-sensory systems.

Ax Qingyue Zhang, Chang Chu, Tianren Peng, Qi Li, Xiangyang Luo, Zhihao Jiang, Shao-Lun Huang 4/9/2026

LoRA-DA: Data-Aware Initialization for Low-Rank Adaptation via Asymptotic Analysis

LoRA-DA framework establishing theoretical foundation for data-aware LoRA initialization using asymptotic analysis for parameter-efficient fine-tuning.

Ax Yuhua Jiang, Shuang Cheng, Yihao Liu, Ermo Hua, Che Jiang, Weigao Sun, Yu Cheng, Feifei Gao, Biqing Qi, Bowen Zhou 4/9/2026

Nirvana: A Specialized Generalist Model With Task-Aware Memory Mechanism

Nirvana specialized generalist model with task-aware memory mechanism combining broad LLM capabilities with domain adaptation.

Ax Mauro Cettolo, Marco Gaido, Matteo Negri, Sara Papi, Luisa Bentivogli 4/9/2026

How to Evaluate Speech Translation with Source-Aware Neural MT Metrics

Neural metrics for speech translation evaluation that incorporate source text information to improve correlation with human judgments.

Ax Zhixiong Zhao, Fangxin Liu, Junjie Wang, Chenyang Guan, Zongwu Wang, Li Jiang, Haibing Guan 4/9/2026

SpecQuant: Spectral Decomposition and Adaptive Truncation for Ultra-Low-Bit LLMs Quantization

SpecQuant framework for ultra-low-bit LLM quantization using spectral decomposition and adaptive truncation for efficient device deployment.

Ax Shihan Cheng, Nilesh Kulkarni, David Hyde, Dmitriy Smirnov 4/9/2026

Less is More: Data-Efficient Adaptation for Controllable Text-to-Video Generation

Data-efficient fine-tuning method for text-to-video diffusion models using sparse synthetic data to add new generative controls.

Ax Zehong Ma, Longhui Wei, Shuai Wang, Shiliang Zhang, Qi Tian 4/9/2026

DeCo: Frequency-Decoupled Pixel Diffusion for End-to-End Image Generation

DeCo framework for efficient pixel-space image generation using frequency-decoupled diffusion transformers.

Ax Jie Li, Hongyi Cai, Mingkang Dong, Muxin Pu, Shan You, Fei Wang, Tao Huang 4/9/2026

Pistachio: Towards Synthetic, Balanced, and Long-Form Video Anomaly Benchmarks

Pistachio synthetic benchmark for video anomaly detection with balanced scene diversity and temporal complexity for autonomous systems.

Ax Chin-Chia Michael Yeh, Uday Singh Saini, Xin Dai, Xiran Fan, Shubham Jain, Yujie Fan, Jiarui Sun, Junpeng Wang, Menghai Pan, Yingtong Dou, Yuzhong Chen, Vineeth Rakesh, Liang Wang, Yan Zheng, Mahashweta Das 4/9/2026

TREASURE: The Visa Payment Foundation Model for High-Volume Transaction Understanding

TREASURE foundation model for transaction understanding in payment networks, enabling anomaly detection and consumer insights at scale.

Ax Run Shao, Ziyu Li, Zhaoyang Zhang, Linrui Xu, Xinran He, Hongyuan Yuan, Bolei He, Yongxing Dai, Yiming Yan, Yijun Chen, Wang Guo, Haifeng Li 4/9/2026

Asking like Socrates: Socrates helps VLMs understand remote sensing images

Socratic questioning framework improving VLM understanding of remote sensing images by addressing pseudo-reasoning and incomplete perception issues.

Ax Huangsen Cao, Qin Mei, Zhiheng Li, Yuxi Li, Zhan Meng, Ying Zhang, Chen Li, Zhimeng Zhang, Xin Ding, Yongwei Wang, Jing Lyu, Fei Wu 4/9/2026

REVEAL: Reasoning-Enhanced Forensic Evidence Analysis for Explainable AI-Generated Image Detection

REVEAL framework for detecting AI-generated images with forensic explainability through structured reasoning rather than post-hoc rationalizations.

Ax Junnan Liu, Hongwei Liu, Songyang Zhang, Kai Chen 4/9/2026

Rectifying LLM Thought from Lens of Optimization

Analysis of chain-of-thought reasoning in LLMs from optimization lens, addressing overthinking and performance issues in long-CoT prompting.

Ax Chihyeon Song, Jaewoo Lee, Jinkyoo Park 4/9/2026

Adaptive Replay Buffer for Offline-to-Online Reinforcement Learning

Adaptive Replay Buffer for offline-to-online reinforcement learning that dynamically balances fixed offline data with new online experiences.

Ax Yuqun Zhang, Yuxuan Zhao, Sijia Chen 4/9/2026

PyFi: Toward Pyramid-like Financial Image Understanding for VLMs via Adversarial Agents

PyFi framework for financial image understanding using vision-language models with adversarial agents and 600K QA dataset organized in reasoning pyramid.

Ax Carla Crivoi, Radu Tudor Ionescu 4/9/2026

Machine Unlearning in the Era of Quantum Machine Learning: An Empirical Study

First empirical study of machine unlearning in hybrid quantum-classical neural networks, adapting classical unlearning methods to variational quantum circuits.

Ax Md Ajoad Hasan, Dipayan Saha, Khan Thamid Hasan, Nashmin Alam, Azim Uddin, Sujan Kumar Saha, Mark Tehranipoor, Farimah Farahmandi 4/9/2026

LAsset: An LLM-assisted Security Asset Identification Framework for System-on-Chip (SoC) Verification

LLM-assisted framework for identifying security assets in SoC designs to improve pre-silicon security verification.

Ax Nelvin Tan, Yaowen Zhang, James Asikin Cheung, Fusheng Liu, Yu-Ching Shih, Dong Yang 4/9/2026

Improved Evidence Extraction and Metrics for Document Inconsistency Detection with LLMs

LLM-based framework for document inconsistency detection with improved evidence extraction capabilities and metrics.

Ax Xiang Li, Xueheng Li, Yu Wang, Xuanhua He, Zhangchi Hu, Weiwei Yu, Chengjun Xie 4/9/2026

Q-Probe: Scaling Image Quality Assessment to High Resolution via Context-Aware Agentic Probing

RL-enhanced MLLM approach for high-resolution image quality assessment using context-aware multi-scale visual probing.

Ax Daixuan Cheng, Shaohan Huang, Yuxian Gu, Huatong Song, Guoxin Chen, Li Dong, Wayne Xin Zhao, Ji-Rong Wen, Furu Wei 4/9/2026

Computer Environments Elicit General Agentic Intelligence in LLMs

Framework virtualizing computer environments as interactive tools to elicit general agentic intelligence capabilities in LLMs.

Ax Jun Xue, Yi Chai, Yanzhen Ren, Jinshen He, Zhiqiang Tang, Zhuolin Yi, Yihuan Huang, Yuankun Xie, Yujie Chen 4/9/2026

Unifying Speech Editing Detection and Content Localization via Prior-Enhanced Audio LLMs

Audio LLM-based approach for detecting speech editing and localizing manipulated content without frame-level supervision.

Ax Shiyu Liu, Xinyi Wen, Zhibin Lan, Ante Wang, Jinsong Su 4/9/2026

Countering the Over-Reliance Trap: Mitigating Object Hallucination for LVLMs via a Self-Validation Framework

Self-validation framework mitigating object hallucination in Large Vision-Language Models via structured analysis.

Ax Powei Chang, Jinpeng Zhang, Bowen Chen, Chenyu Wang, Chenlu Guo, Yixing Zhang, Yukang Gao, JianXiang Xiang, Yue Gao, Chaoqun Sun, Yiyi Chen, Dongying Kong 4/9/2026

SPICE: Submodular Penalized Information-Conflict Selection for Efficient Large Language Model Training

Submodular-based data selection for efficient LLM instruction tuning by addressing gradient conflicts in Fisher information.

Ax Basile Terver, Randall Balestriero, Megi Dervishi, David Fan, Quentin Garrido, Tushar Nagarajan, Koustuv Sinha, Wancong Zhang, Mike Rabbat, Yann LeCun, Amir Bar 4/9/2026

A Lightweight Library for Energy-Based Joint-Embedding Predictive Architectures

Open-source library for learning representations and world models using Joint-Embedding Predictive Architectures.

Ax J Rosser, Robert Kirk, Edward Grefenstette, Jakob Foerster, Laura Ruis 4/9/2026

Infusion: Shaping Model Behavior by Editing Training Data via Influence Functions

Framework using influence functions to craft training data perturbations that induce targeted behavior changes in models.

Ax Paul Tschisgale, Peter Wulff 4/9/2026

Daily and Weekly Periodicity in Large Language Model Performance and Its Implications for Research

Longitudinal study revealing daily and weekly performance variations in LLMs, impacting research reproducibility and reliability.

Ax Chien-Ping Lu 4/9/2026

Modernizing Amdahl's Law: How AI Scaling Laws Shape Computer Architecture

Reformulates Amdahl's Law for modern heterogeneous AI systems with constrained resource allocation across diverse hardware.

Ax Lekshmi P, Neha Karanjkar 4/9/2026

On Integrating Resilience and Human Oversight into LLM-Assisted Modeling Workflows for Digital Twins

Proposes design principles for integrating resilience and human oversight into LLM-assisted digital twin modeling workflows.

Ax Xianyong Xu, Yuanjun Zuo, Zhihong Huang, Yihan Qin, Haoxian Xu, Leilei Du, Haotian Wang 4/9/2026

MR-ImagenTime: Multi-Resolution Time Series Generation through Dual Image Representations

MR-CDM: Multi-resolution time series forecasting framework using hierarchical decomposition and diffusion-based generation.

Ax Aichen Cai, Anmeng Zhang, Anyu Li, Bo Zhang, Bohua Cai, Chang Li, Changjian Jiang, Changkai Lu, Chao Xue, Chaocai Liang, Cheng Zhang, Dongkai Liu, Fei Wang, Guoqiang Huang, Haijian Ke, Han Lin, Hao Wang, Ji Miao, Jiacheng Zhang, Jialong Shi, Jifeng Zhu, Jingjing Qian, Junhui Luo, Junwu Xiong, Lam So, Liang Huang, Ming Ke, Mingyang Li, Panfeng Shi, Peng Hao, Qi Wang, Qian Lai, Qiaoqiao Yuan, Qingyu Yin, Qiong Cao, Qixiang Wang, Rongcheng Bian, Rongduo Han, Shaoqiang Zheng, Shi Hu, Shi Suo, Shijie Ren, Shijin Zhang, Shiying Fan, Shuai Xie, Tianyi Zhang, Wei Liu, Wentao Tan, Xianghan Meng, Xiaodong He, Xing Pan, Xiran Wang, Xuyang Peng, Ya Zhang, Yang Liu, Yangyang Duan, Yanxu Chen, Yicheng Gong, Yidan Huang, Yifei Liu, Yinhao Bai, Yongqiang Liu, Yuesong Zhang, Yuqi Zhang, Zerui Xie, Zhenfang Wang, Zhennan Shen, Zheyuan Liu, Zhuwei Zeng 4/9/2026