Isolater - Feed

Ax Indraveni Chebolu, Arnab Mallick, Harmesh Rana 4/13/2026

SPEAR: An Engineering Case Study of Multi-Agent Coordination for Smart Contract Auditing

Multi-agent framework for smart contract auditing using specialized agents for planning, execution, and recovery with coordination protocols.

Ax Eun Cheol Choi, Lindsay E. Young, Emilio Ferrara 4/13/2026

Overstating Attitudes, Ignoring Networks: LLM Biases in Simulating Misinformation Susceptibility

Study demonstrating LLM biases when simulating misinformation susceptibility, showing models overstate attitudes and ignore network effects present in humans.

Ax Prerna Ravi, Car\'umey Stevens, Beatriz Flamia Azevedo, Jasmine David, Brandon Hanks, Hal Abelson, Grace Lin, Emma Anderson 4/13/2026

Exploring Teachers' Perspectives on Using Conversational AI Agents for Group Collaboration

Qualitative study of 33 K12 teachers' perspectives on using conversational AI agents to scaffold group collaboration in classrooms.

Ax Zhaoyang Zhang, Shuli Jiang, Yantao Shen, Yuting Zhang, Dhananjay Ram, Shuo Yang, Zhuowen Tu, Wei Xia, Stefano Soatto 4/13/2026

Reinforcement-aware Knowledge Distillation for LLM Reasoning

Knowledge distillation method for distilling RL-trained LLMs with chain-of-thought reasoning into smaller student models while preserving reasoning capabilities.

Ax Yuan Wu, Zongxian Yang, Jiayu Qian, Songpan Gao, Guanxing Chen, Qiankun Li, Yu-An Huang, Zhi-An Huang 4/13/2026

Better Eyes, Better Thoughts: Why Vision Chain-of-Thought Fails in Medicine

Analysis showing chain-of-thought prompting underperforms direct answering in medical vision-language models due to perception bottlenecks in domain-specific tasks.

Ax Zhexi Lian, Haoran Wang, Xuerun Yan, Weimeng Lin, Xianhong Zhang, Yongyu Chen, Jia Hu 4/13/2026

Fine-tuning is Not Enough: A Parallel Framework for Collaborative Imitation and Reinforcement Learning in End-to-end Autonomous Driving

Parallel framework combining imitation and reinforcement learning for autonomous driving, addressing limitations of sequential fine-tuning approaches.

Ax Omkar Patil, Ondrej Biza, Thomas Weng, Karl Schmeckpeper, Wil Thomason, Xiaohan Zhang, Robin Walters, Nakul Gopalan, Sebastian Castro, Eric Rosen 4/13/2026

You've Got a Golden Ticket: Improving Generative Robot Policies With A Single Noise Vector

Method to improve pretrained generative robot policies by replacing sampled noise with optimized constant noise vectors for downstream reward optimization.

Ax Mengxian Lyu, Cheng Peng, Ziyi Chen, Mengyuan Zhang, Jieting Li Lu, Yonghui Wu 4/13/2026

Improving Automatic Summarization of Radiology Reports through Mid-Training of Large Language Models

Mid-training adaptation strategy for LLMs to improve automatic summarization of radiology reports, exploring domain-specific pre-training approaches.

Ax Robert Aufschl\"ager, Jakob Folz, Gautam Savaliya, Manjitha D Vidanalage, Michael Heigl, Martin Schramm 4/13/2026

Towards Context-Aware Image Anonymization with Multi-Agent Reasoning

CAIAMAR: multi-agent framework for context-aware image anonymization in street-level imagery using agentic reasoning.

Ax Haochuan Kevin Wang, Zechen Zhang 4/13/2026

Kill-Chain Canaries: Stage-Level Tracking of Prompt Injection Across Attack Surfaces and Model Safety Tiers

Kill-chain canary methodology for tracking prompt injection attacks across multi-agent LLM systems with stage-level diagnostics.

Ax Hita Kambhamettu, Will Crichton, Sean Welleck, Harrison Goldstein, Andrew Head 4/13/2026

Explorable Theorems: Making Written Theorems Explorable by Grounding Them in Formal Representations

System for making mathematical theorems interactive by grounding LLM-generated explanations in formal representations enabling execution and stepping.

Ax Myra Cheng, Isabel Sieh, Humishka Zope, Sunny Yu, Lujain Ibrahim, Aryaman Arora, Jared Moore, Desmond Ong, Dan Jurafsky, Diyi Yang 4/13/2026

Verbalizing LLMs' assumptions to explain and control sycophancy

Framework for eliciting and verbalizing LLM assumptions to explain and mitigate sycophantic behavior in model outputs.

Ax Yi Zhou 4/13/2026

From Paper to Program: Accelerating Quantum Many-Body Algorithm Development via a Multi-Stage LLM-Assisted Workflow

Multi-stage LLM-assisted workflow for scientific algorithm development separating theory extraction, formal specification, and code generation.

Ax Cheol Woo Kim, Jai Moondra, Roozbeh Nahavandi, Andrew Perrault, Milind Tambe, Swati Gupta 4/13/2026

Many Preferences, Few Policies: Towards Scalable Language Model Personalization

Method for LLM personalization using a small portfolio of models capturing diverse user preferences without per-user models.

Ax Zequn Chen, Wesley J. Marrero 4/13/2026

Boosted Distributional Reinforcement Learning: Analysis and Healthcare Applications

Distributional reinforcement learning approach for decision-making in healthcare, accounting for uncertainty across heterogeneous populations.

Ax Jingwei Zuo, Xinze Feng, Zien Liu, Kaijian Wang, Fanjiang Ye, Ye Cao, Zhuang Wang, Yuke Wang 4/13/2026

ALTO: Adaptive LoRA Tuning and Orchestration for Heterogeneous LoRA Training Workloads

ALTO: system for adaptive LoRA hyperparameter tuning and orchestration across heterogeneous LLM fine-tuning workloads in multi-tenant environments.

Ax Yiquan Wu, Yuhang Liu, Yifei Liu, Ang Li, Siying Zhou, Kun Kuang, Fei Wu 4/13/2026

WisdomInterrogatory (LuWen): An Open-Source Legal Large Language Model Technical Report

WisdomInterrogatory (LuWen): open-source Chinese legal language model built on Baichuan foundation model for legal domain applications.

Ax Xue Qin, Simin Luan, John See, Cong Yang, Zhijun Li 4/13/2026

Governed Capability Evolution for Embodied Agents: Safe Upgrade, Compatibility Checking, and Runtime Rollback for Embodied Capability Modules

System for safe capability evolution in embodied agents with compatibility checking and runtime rollback mechanisms.

Ax Seungjae Moon, Seunghyun Oh, Youngmin Ro 4/13/2026

OV-Stitcher: A Global Context-Aware Framework for Training-Free Open-Vocabulary Semantic Segmentation

Training-free open-vocabulary semantic segmentation framework (OV-Stitcher) leveraging pretrained vision-language models without additional training.

Ax Juwei Yue, Chuanrui Hu, Jiawei Sheng, Zuyi Zhou, Wenyuan Zhang, Tingwen Liu, Li Guo, Yafeng Deng 4/13/2026

HyperMem: Hypergraph Memory for Long-Term Conversations

HyperMem: hypergraph-based memory architecture for conversational agents enabling long-term context tracking and high-order associations.

Ax Yunsong Zhou, Hangxu Liu, Xuekun Jiang, Xing Shen, Yuanzhen Zhou, Hui Wang, Baole Fang, Yang Tian, Mulin Yu, Qiaojun Yu, Li Ma, Hengjie Li, Hanqing Wang, Jia Zeng, Jiangmiao Pang 4/13/2026

SIM1: Physics-Aligned Simulator as Zero-Shot Data Scaler in Deformable Worlds

Physics-aligned simulator (SIM1) for generating synthetic data in deformable object robotic manipulation tasks.

Ax Ruiyao Xu, Kaize Ding 4/13/2026

GNN-as-Judge: Unleashing the Power of LLMs for Graph Learning with GNN Feedback

Framework combining LLMs with graph neural networks for text-attributed graph learning in low-resource settings using GNN feedback.

Ax Ali Slim, Haydar Hamieh, Jawad Kotaich, Yehya Ghosn, Mahdi Chehimi, Ammar Mohanna, Hasan Abed Al Kader Hammoud, Bernard Ghanem 4/13/2026

QuanBench+: A Unified Multi-Framework Benchmark for LLM-Based Quantum Code Generation

QuanBench+ unified benchmark for LLM quantum code generation across Qiskit, PennyLane, Cirq with 42 aligned executable tasks.

Ax Pavel Golikov, Evgenii Opryshko, Gennady Pekhimenko, Mark C. Jeffrey 4/13/2026

Robust Reasoning Benchmark

Benchmark evaluating robustness of LLM reasoning with 14 perturbation techniques applied to mathematical reasoning tasks.

Ax Yeping Jin, Jiaming Hu, Ioannis Ch. Paschalidis 4/13/2026

Distributionally Robust Token Optimization in RLHF

DRTO combines token-level RLHF with distributional robustness to improve LLM resilience to input perturbations and formatting changes.

Ax Phong Lam, Ha-Linh Nguyen, Thu-Trang Nguyen, Son Nguyen, Hieu Dinh Vo 4/13/2026

Structured Exploration and Exploitation of Label Functions for Automated Data Annotation

Automated label function generation for data annotation using LLMs with structured exploration-exploitation strategy.

Ax Abdulrahman Albaiz, Fathi Amsaad 4/13/2026

Fully Autonomous Z-Score-Based TinyML Anomaly Detection on Resource-Constrained MCUs Using Power Side-Channel Data

TinyML Z-score anomaly detection system running on resource-constrained microcontrollers using power side-channel data.

Ax Chuxu Song, Zhencan Peng, Jiuqi Wei, Chuanhui Yang 4/13/2026

CSAttention: Centroid-Scoring Attention for Accelerating LLM Inference

CSAttention: sparse attention mechanism for accelerating LLM inference by reducing KV-cache bottlenecks through centroid-scoring without retraining.

Ax Matthew DosSantos DiSorbo, Harang Ju 4/13/2026

Act or Escalate? Evaluating Escalation Behavior in Automation with Language Models

Framework evaluating when LLMs should act versus escalate decisions using uncertainty estimation across five real-world domains.

Ax Brendan R. Hogan, Xiwen Chen, James T. Wilson, Kashif Rasul, Adel Boyarsky, Thomas Kamei, Anderson Schneider, Yuriy Nevmyvaka 4/13/2026

AlphaLab: Autonomous Multi-Agent Research Across Optimization Domains with Frontier LLMs

AlphaLab autonomous research system using frontier LLMs as agents to automate full experimental cycles in optimization domains without human intervention.

Ax Ivo Nowak 4/13/2026

StructRL: Recovering Dynamic Programming Structure from Learning Dynamics in Distributional Reinforcement Learning

StructRL: recovers dynamic programming structure from distributional RL learning dynamics. Bridges data-driven and structured approaches for stable learning.

Ax Yesmine Abdennadher, Philip N. Garner 4/13/2026

Practical Bayesian Inference for Speech SNNs: Uncertainty and Loss-Landscape Smoothing

Bayesian inference for spiking neural networks in speech processing. Explores weight uncertainty and loss landscape smoothing for temporal tasks.

Ax Yongchan Chun, Chanhee Park, Jeongho Yoon, Jaehyung Seo, Heuiseok Lim 4/13/2026

Evidential Transformation Network: Turning Pretrained Models into Evidential Models for Post-hoc Uncertainty Estimation

Evidential Transformation Network: adapts pretrained models for post-hoc uncertainty estimation. Efficient alternative to ensembles/MC dropout for deployed models.

Ax Rahul D Ray, Utkarsh Srivastava 4/13/2026

VOLTA: The Surprising Ineffectiveness of Auxiliary Losses for Calibrated Deep Learning

VOLTA: benchmark comparing uncertainty quantification methods for deep learning. Evaluates 10 UQ baselines across modalities and distribution shifts.

Ax Ramakrishnan Krishnamurthy, Arpit Agarwal, Lakshminarayanan Subramanian, Maximilian Nickel 4/13/2026

Creator Incentives in Recommender Systems: A Cooperative Game-Theoretic Approach for Stable and Fair Collaboration in Multi-Agent Bandits

Game-theoretic analysis of creator incentives in multi-agent recommender systems. Cooperative game formulation for fair collaboration in bandit problems.

Ax Maxim Ostroukhov, Ruslan Mikhailov, Vladimir Iashin, Artem Sokolov, Andrei Akshonov, Vitaly Protasov, Dmitrii Beloborodov, Vince Mullin, Roman Yokunda Enzmann, Georgios Kolovos, Jason Renders, Pavel Nesterov, Anton Repushko 4/13/2026

PRAGMA: Revolut Foundation Model

PRAGMA: foundation models for banking event sequences. Transformer-based architecture with self-supervised pretraining on financial transaction data.

Ax Fengwei Teng, Jinyi Bai, Xinhao Yao, Demi Ruohan Wang, Jiahao Zhao, Zhijiang Guo 4/13/2026

Skip-Connected Policy Optimization for Implicit Advantage

Skip-Connected Policy Optimization (SKPO) for reinforcement learning with reasoning tasks. Improves upon GRPO by addressing high-variance advantage estimation.

Ax Nan Huang, Xiaoxiao Zhou, Junxia Cui, Mario Tapia-Pacheco, Tiffany Amariuta, Yang Li, Jingbo Shang 4/13/2026

EvoLen: Evolution-Guided Tokenization for DNA Language Model

EvoLen: evolution-guided tokenization approach for DNA language models. Addresses fundamental tokenization design challenges in biological sequence modeling.

Ax Charles Arnal, Vivien Cabannes, Taco Cohen, Julia Kempe, Remi Munos 4/13/2026

Efficient RL Training for LLMs with Experience Replay

Experience replay for LLM post-training RL formalizing optimal buffer design as trade-off between sample efficiency and data freshness.

Ax Tiejin Chen, Huaiyuan Yao, Jia Chen, Evangelos E. Papalexakis, Hua Wei 4/13/2026

Every Response Counts: Quantifying Uncertainty of LLM-based Multi-Agent Systems through Tensor Decomposition

Tensor decomposition method quantifying uncertainty in LLM-based multi-agent systems accounting for communication and role dependencies.

Ax Diyi Hu, Bhaskar Krishnamachari 4/13/2026

Wireless Communication Enhanced Value Decomposition for Multi-Agent Reinforcement Learning

CLOVER framework for multi-agent RL cooperation conditioning value decomposition on realistic wireless communication graphs.

Ax Hananel Hazan, Yanbo Zhang, Benedikt Hartl, Michael Levin 4/13/2026

A Little Rank Goes a Long Way: Random Scaffolds with LoRA Adapters Are All You Need

LottaLoRA training paradigm showing frozen random backbones with trained LoRA adapters recover 96-100% performance across diverse tasks.

Ax Mingjie Hu, Siyang Gao, Jian-qiang Hu, Enlu Zhou 4/13/2026

Adaptive Simulation Experiment for LLM Policy Optimization

Adaptive simulation experiment framework using pairwise comparisons to optimize LLM policies for operations management tasks.

Ax Zhaolin Gao (Sid), Yu (Sid), Wang, Bo Liu, Thorsten Joachims, Kiant\'e Brantley, Wen Sun 4/13/2026

$p1$: Better Prompt Optimization with Fewer Prompts

Prompt optimization method decomposing reward variance into response and prompt variance to identify task amenability to optimization.

Ax Yashodhan D. Hakke, Almuatazbellah M. Boker, Lamine Mili, Michael von Spakovsky, Hoda Eldardiry 4/13/2026

Alleviating Community Fear in Disasters via Multi-Agent Actor-Critic Reinforcement Learning

Multi-agent actor-critic reinforcement learning for disaster resilience controlling power, communication, and emergency response systems.

Ax Mehran Taghian, Yunke Peng, Xing Huang, Yao Wang, Yaoyuan Wang, Wei Guo, Yuanyong Luo, Tianchi Hu, Junsong Wang, Xin Wang, Hu Liu, Yu Cheng, Ziwei Yu, Hongliang Li, Mehdi Rahimifar, Lei Yan, Xuefei Wang, Zhuang Ma, Lei Liu, Hui Yu, Anandharaju Durai Raju, Hoang Le, Hei Yi Mak, Tanzila Rahman, Shadan Golestan 4/13/2026