Ax Bingbing Wen, Sirajul Salekin, Feiyang Kang, Bill Howe, Lucy Lu Wang, Javier Movellan, Manjot Bilkhu 4/17/2026

MixAtlas: Uncertainty-aware Data Mixture Optimization for Multimodal LLM Midtraining

MixAtlas: data mixture optimization method for multimodal LLM training with uncertainty-aware reweighting.

Ax Pu Cheng, Juncheng Liu, Yunshen Long 4/17/2026

PolyBench: Benchmarking LLM Forecasting and Trading Capabilities on Live Prediction Market Data

PolyBench: multimodal benchmark for evaluating LLM forecasting and trading on live prediction market data.

Ax Hanchen David Wang, Diego Manzanas Lopez, Preston K. Robinette, Ipek Oguz, Taylor T. Johnson, Meiyi Ma 4/17/2026

Towards Verified and Targeted Explanations through Formal Methods

Formal methods approach to verified neural network explanations with mathematical guarantees. Combines XAI with formal verification for safety-critical domains.

Ax Deep Shah, Sanket Badhe, Nehal Kathrotia, Priyanka Tiwari 4/17/2026

CROP: Token-Efficient Reasoning in Large Language Models via Regularized Prompt Optimization

CROP: Automatic prompt optimization method with regularization for token efficiency in LLM reasoning. Balances accuracy against inference cost and latency.

Ax Richard Wai Cheung Chan, Shanru Lin, Ya-nan Ma, Hao Chen, Liangjun Jiang, Wenqi Fan 4/17/2026

PriHA: A RAG-Enhanced LLM Framework for Primary Healthcare Assistant in Hong Kong

PriHA: RAG-enhanced LLM framework for primary healthcare assistant in Hong Kong, grounding clinical guidelines to improve access and reduce healthcare costs.

Ax Aizierjiang Aiersilan, Mohamad Koubeissi 4/17/2026

Neuro-Oracle: A Trajectory-Aware Agentic RAG Framework for Interpretable Epilepsy Surgical Prognosis

Neuro-Oracle: Trajectory-aware agentic RAG framework for epilepsy surgical prognosis using 3D Siamese contrastive encoding and longitudinal MRI analysis.

Ax Koushik Chakraborty, Koyel Guha 4/17/2026

Knowledge Graph RAG: Agentic Crawling and Graph Construction in Enterprise Documents

Knowledge graph RAG with agentic recursive crawling for enterprise documents. Addresses hierarchical information capture in complex document ecosystems.

Ax Afshan Hashmi 4/17/2026

Adaptive Query Routing: A Tier-Based Framework for Hybrid Retrieval Across Financial, Legal, and Medical Documents

Tier-based adaptive routing framework for hybrid RAG across financial, legal, medical documents. Compares vector-based agentic vs hierarchical reasoning approaches.

Ax Ashmi Banerjee, Adithi Satish, Wolfgang W\"orndl, Yashar Deldjoo 4/17/2026

TRACE: A Conversational Framework for Sustainable Tourism Recommendation with Agentic Counterfactual Explanations

TRACE: Multi-agent LLM framework using counterfactual explanations for sustainable tourism recommendations. Modular orchestrator-worker architecture.

Ax Sohyun An (Meta Superintelligence Labs, UCLA), Hayeon Lee (Meta Superintelligence Labs), Shuibenyang Yuan (Meta Superintelligence Labs), Chun-cheng Jason Chen (Meta Superintelligence Labs), Cho-Jui Hsieh (UCLA), Vijai Mohan (Meta Superintelligence Labs), Alexander Min (Meta Superintelligence Labs) 4/17/2026

FRESCO: Benchmarking and Optimizing Re-rankers for Evolving Semantic Conflict in Retrieval-Augmented Generation

FRESCO benchmark for evaluating re-rankers in RAG systems under evolving semantic information. Addresses temporal staleness in LLM-grounded retrieval.

Ax Jiacheng Liu, Xiaohan Zhao, Xinyi Shang, Zhiqiang Shen 4/17/2026

Dive into Claude Code: The Design Space of Today's and Future AI Agent Systems

Analysis of Claude Code agentic architecture via TypeScript source code, compared with OpenClaw open-source AI agent system. Identifies design patterns and human values.

Ax Mohammad Nasir Uddin, Md Munna Aziz 4/17/2026

Shapley Value-Guided Adaptive Ensemble Learning for Explainable Financial Fraud Detection with U.S. Regulatory Compliance Validation

Shapley value-based ensemble learning for explainable fraud detection with regulatory compliance. ML research on interpretability in financial domain.

Ax Sourav Ganguly, Kartik Pandit, Arnob Ghosh 4/17/2026

Optimistic Policy Learning under Pessimistic Adversaries with Regret and Violation Guarantees

Optimistic policy learning under adversarial exogenous factors with regret and constraint violation guarantees.

Ax Wentao Hu, Yanbo Zhai, Xiaohui Hu, Mingkuan Zhao, Shanhong yu, Xue Liu, Kaidong Yu, Shuangyong Song, Xuelong Li 4/17/2026

Awakening Dormant Experts:Counterfactual Routing to Mitigate MoE Hallucinations

Counterfactual routing method to mitigate hallucinations in sparse Mixture-of-Experts models by activating dormant experts on long-tail knowledge.

Ax To Eun Kim, Alireza Salemi, Hamed Zamani, Fernando Diaz 4/17/2026

Evaluation of Agents under Simulated AI Marketplace Dynamics

Framework for evaluating agent performance under simulated marketplace dynamics with user switching, routing, and competitive constraints.

Ax Zhuofeng Li, Yi Lu, Dongfu Jiang, Haoxiang Zhang, Yuyang Bai, Chuan Li, Yu Wang, Shuiwang Ji, Jianwen Xie, Yu Zhang 4/17/2026

ReviewGrounder: Improving Review Substantiveness with Rubric-Guided, Tool-Integrated Agents

ReviewGrounder: rubric-guided, tool-integrated LLM agents for generating substantive peer review feedback grounded in existing work.

Ax Yangyue Wang, Harshvardhan Sikka, Yash Mathur, Tony Zhou, Jinu Nyachhyon, Pranav Guruprasad 4/17/2026

GUI-Perturbed: Domain Randomization Reveals Systematic Brittleness in GUI Grounding Models

GUI-Perturbed framework reveals brittleness in GUI grounding models through controlled perturbations of visual scenes and instructions.

Ax Haoran Xu, Kaiwen Hu, Somayeh Sojoudi, Amy Zhang 4/17/2026

Reinforcement Learning via Value Gradient Flow

Value gradient flow method for behavior-regularized reinforcement learning in LLM finetuning without reparameterized policy gradients.

Ax Junzhe Wang, Zhiheng Xi, yajie yang, Hao Luo, Shihan Dou, Tao Gui, Qi Zhang 4/17/2026

Enhancing LLM-based Search Agents via Contribution Weighted Group Relative Policy Optimization

Contribution weighted group relative policy optimization for training LLM-based search agents with improved credit assignment and value estimation.

Ax Guillermo Valverde, Igor Garc\'ia-Olaizola, Giannicola Scarpa, Alejandro Pozas-Kerstjens 4/17/2026

Quantum-inspired tensor networks in machine learning models

Tensor networks from quantum physics integrated into machine learning models for compressed representations of complex correlations.

Ax Francesco Andrea Causio, Vittorio De Vita, Olivia Riccomi, Michele Ferramola, Federico Felizzi, Antonio Cristiano, Lorenzo De Mori, Chiara Battipaglia, Melissa Sawaya, Luigi De Angelis, Marcello Di Pumpo, Alessandra Piscitelli, Pietro Eric Risuleo, Alessia Longo, Giulia Vojvodic, Mariapia Vassalli, Bianca Destro Castaniti, Nicol\`o Scarsi, Manuel Del Medico 4/17/2026

EuropeMedQA Study Protocol: A Multilingual, Multimodal Medical Examination Dataset for Language Model Evaluation

EuropeMedQA: multilingual, multimodal medical examination dataset for evaluating LLM performance on non-English diagnostic tasks.

Ax Gabriel Pimenta de Freitas Cardoso, Caio Lucas da Silva Chacon, Jonas Felipe da Fonseca Oliveira, Paulo Henrique de Medeiros Araujo 4/17/2026

DharmaOCR: Specialized Small Language Models for Structured OCR that outperform Open-Source and Commercial Baselines

DharmaOCR: specialized small language models for structured OCR with benchmark covering printed, handwritten, and legal documents.

Ax Salem Radey, Jack West, Kassem Fawaz 4/17/2026

Challenges and Future Directions in Agentic Reverse Engineering Systems

Study of agentic LLM systems for binary reverse engineering, examining limitations in obfuscation, timing, and unique architecture handling.

Ax Bar Alon, Itamar Zimerman, Lior Wolf 4/17/2026

Faithfulness Serum: Mitigating the Faithfulness Gap in Textual Explanations of LLM Decisions via Attribution Guidance

Attribution guidance method to improve faithfulness of textual explanations from LLMs, reducing gap between convincing and accurate rationales.

Ax Pushpa Kumar Balan, Aijing Feng 4/17/2026

Mamba-SSM with LLM Reasoning for Biomarker Discovery: Causal Feature Refinement via Chain-of-Thought Gene Evaluation

Uses Mamba SSM and LLM chain-of-thought reasoning to filter confounders in biomarker discovery from RNA-seq data.

Ax Tianhao Qian 4/17/2026

Tight Sample Complexity Bounds for Best-Arm Identification Under Bounded Systematic Bias

Sample complexity bounds for best-arm identification in autonomous reasoning with safety guarantees for LLM-based node expansion in planning.

Ax Pratyay Banerjee, Masud Moshtaghi, Shivashankar Subramanian, Amita Misra, Ankit Chadha 4/17/2026

APEX-MEM: Agentic Semi-Structured Memory with Temporal Reasoning for Long-Term Conversational AI

APEX-MEM: conversational memory system for LLMs using property graphs with temporal reasoning and entity-centric framework to improve long-term context retention.

Ax Akshay Paruchuri, Ishan Chatterjee, Henry Fuchs, Ehsan Adeli, Piotr Didyk 4/17/2026

The Cost of Language: Centroid Erasure Exposes and Exploits Modal Competition in Multimodal Language Models

Analysis of modal dependence in multimodal language models showing text centroid structure is 4x more critical than visual structure for performance.

Ax Xue Wu, Shengting Cao, Jiaqi Gong 4/17/2026

SatBLIP: Context Understanding and Feature Identification from Satellite Imagery with Vision-Language Learning

SatBLIP: satellite-specific vision-language model for rural context understanding and social vulnerability index prediction from satellite imagery.

Ax Noureddine Kermiche 4/17/2026

Modular Continual Learning via Zero-Leakage Reconstruction Routing and Autonomous Task Discovery

Modular continual learning architecture addressing catastrophic forgetting through task-specific experts, gatekeeper routing, and simultaneous pipeline training.

Ax Qi Zhang, Dawei Wang, Shaofeng Zou 4/17/2026

Step-level Denoising-time Diffusion Alignment with Multiple Objectives

Step-level diffusion model alignment approach using reinforcement learning to balance multiple human preference objectives beyond single reward optimization.

Ax Dongxin Guo, Jikun Wu, Siu-Ming Yiu 4/17/2026

Coalition Formation in LLM Agent Networks: Stability Analysis and Convergence Guarantees

First formal framework analyzing coalition formation in multi-agent LLM systems using hedonic game theory with stability guarantees and convergence proofs.

Ax Hyunkyung Park, Arkaitz Zubiaga 4/17/2026

BiCon-Gate: Consistency-Gated De-colloquialisation for Dialogue Fact-Checking

BiCon-Gate system for dialogue fact-checking using de-colloquialization and consistency gates to handle informal language in multi-turn conversations.

Ax Aodi Wu, Haodong Han, Xubo Luo, Ruisuo Wang, Shan He, Xue Wan 4/17/2026

SpaceMind: A Modular and Self-Evolving Embodied Vision-Language Agent Framework for Autonomous On-orbit Servicing

SpaceMind: modular vision-language agent framework for autonomous orbital servicing using skill modules, dynamic routing, and Model Context Protocol tools.

Ax Vekil Bekmyradov, Noah C. P\"utz, Thomas Bartz-Beielstein 4/17/2026

LLMs taking shortcuts in test generation: A study with SAP HANA and LevelDB

Research examining how LLMs rely on shallow heuristics and memorization rather than genuine reasoning in automated test generation for software like SAP HANA and LevelDB.

Ax Sang-Il Han 4/17/2026

Hierarchical vs. Flat Iteration in Shared-Weight Transformers

HRM-LM: Empirical study of hierarchical shared-weight recurrence versus independent Transformer layers for language model representation.

Ax Nikola Zubi\'c, Qian Li, Yuyi Wang, Davide Scaramuzza 4/17/2026

On the Expressive Power and Limitations of Multi-Layer SSMs

Theoretical analysis of multi-layer SSM expressiveness, showing fundamental compositional limitations and CoT benefits.

Ax Rongyao Wang, Veronica Liesaputra, Zhiyi Huang 4/17/2026

NewsTorch: A PyTorch-based Toolkit for Learner-oriented News Recommendation

PyTorch toolkit for news recommendation research supporting learners with conceptual and practical experience.

Ax Hugo O'Connor 4/17/2026

CBCL: Safe Self-Extending Agent Communication

Agent communication language with constrained extensibility enabling safe heterogeneous agent coordination across domains.

Ax Sazzadul Islam, Tasnim Tabassum, Hao Zheng 4/17/2026

VeriGraphi: A Multi-Agent Framework of Hierarchical RTL Generation for Large Hardware Designs

Multi-agent hierarchical framework using LLMs to generate synthesizable Verilog for large hardware designs with structural reasoning.

Ax Yiqun Sun, Pengfei Wei, Lawrence B. Hsieh 4/17/2026

Don't Retrieve, Navigate: Distilling Enterprise Knowledge into Navigable Agent Skills for QA and RAG

Corpus2Skill distills document corpora into navigable hierarchical skill directories for LLM agents performing QA and RAG tasks.

Ax Cheng Lu, Mengxin Wang, Dennis J. Zhang, Heng Zhang 4/17/2026

Generative Augmented Inference

Framework leveraging LLM outputs as auxiliary data for operations management and causal inference under distribution shift.

Ax Shreesha G. Bhat, Tony Hong, Michael Noguera, Ramnatthan Alagappan, Aishwarya Ganesan 4/17/2026

AgileLog: A Forkable Shared Log for Agents on Data Streams

Shared log system enabling AI agents to reason over streaming data without performance interference in data-streaming architectures.

Ax Yitong Shou, Manhao Guan 4/17/2026

Mechanistic Decoding of Cognitive Constructs in LLMs

Reverse-engineering framework to mechanistically decode complex emotional and cognitive constructs within LLM internals.

Ax Yian Wang, Yuen Chen, Agam Goyal, Hari Sundaram 4/17/2026

CausalDetox: Causal Head Selection and Intervention for Language Model Detoxification

Framework using causal intervention on attention heads to reduce toxic content generation in LLMs while maintaining quality.

Ax Meng Chen, Kun Wang, Li Lu, Jiaheng Zhang, Tianwei Zhang 4/17/2026

Hijacking Large Audio-Language Models via Context-Agnostic and Imperceptible Auditory Prompt Injection

Security vulnerabilities in large audio-language models exposed through imperceptible adversarial audio injection attacks.

Ax Sumit Mukherjee, Juan Shu, Nairwita Mazumder, Tate Kernell, Celena Wheeler, Shannon Hastings, Chris Sidey-Gibbons 4/17/2026

Retrieve, Then Classify: Corpus-Grounded Automation of Clinical Value Set Authoring

Retrieval-augmented approach for automating clinical value set authoring by grounding LLM generation in standardized medical vocabularies.

Ax Sanidhya Vijayvargiya, Vijay Viswanathan, Graham Neubig 4/17/2026

Asking What Matters: Reward-Driven Clarification for Software Engineering Tasks

Studies clarification question generation in software engineering tasks, identifying which missing information types most affect task success with LLM assistants.

Ax Yuseon Choi, Jingu Lee, Jungjun Oh, Sunjoo Whang, Byeongcheol Kim, Minsung Kim, Hoi-Jun Yoo, Sangjin Kim 4/17/2026

ELMoE-3D: Leveraging Intrinsic Elasticity of MoE for Hybrid-Bonding-Enabled Self-Speculative Decoding in On-Premises Serving

ELMoE-3D optimizes Mixture-of-Experts model serving via speculative decoding and memory-centric architectures for on-premises deployment.

Ax Geonhui Jang, Dongyoon Han, YoungJoon Yoo 4/17/2026

StoryCoder: Narrative Reformulation for Structured Reasoning in LLM Code Generation

StoryCoder framework transforms fragmented problem conditions into coherent narratives to improve LLM code generation through better structured reasoning.