Isolater - Feed

Ax Fabian A Mikulasch, Friedemann Zenke 5/6/2026

Understanding Self-Supervised Learning via Latent Distribution Matching

Provides unifying theoretical framework for self-supervised learning via latent distribution matching principle.

Ax Hao Zhou, Rui Zhang, Han Wan, Hao Sun 5/6/2026

PerFlow: Physics-Embedded Rectified Flow for Efficient Reconstruction and Uncertainty Quantification of Spatiotemporal Dynamics

Combines physics constraints with rectified flow for reconstructing spatiotemporal PDE-governed fields from sparse measurements.

Ax Jorge L. Ruiz Williams 5/6/2026

HeadQ: Model-Visible Distortion and Score-Space Correction for KV-Cache Quantization

Presents HeadQ, a KV-cache quantization method optimizing for model-visible distortion in LLM inference.

Ax Moritz Grillo, Guido Mont\'ufar 5/6/2026

Most ReLU Networks Admit Identifiable Parameters

Studies parameter identifiability in deep ReLU networks using weighted polyhedral complexes framework.

Ax Zhiqi Li, Duowen Chen, Yuchen Sun, Bo Zhu 5/6/2026

A Few-Step Generative Model on Cumulative Flow Maps

Proposes few-step generative modeling framework using cumulative flow maps for probability space transport, inspired by physical dynamics.

Ax Maximilian Nothnagel, Bernhard C. Geiger 5/6/2026

Information Plane Analysis of Binary Neural Networks

Information plane analysis of binary neural networks using discrete activations to overcome MI estimation challenges in deep network training dynamics.

Ax Jiaxi Li, Lu Yin, Li Shen, Jinjin Xu, Yuhui Liu, Wenwu Wang, Shiwei Liu, Xilu Wang 5/6/2026

ELAS: Efficient Pre-Training of Low-Rank Large Language Models via 2:4 Activation Sparsity

ELAS combines low-rank training and 2:4 activation sparsity for efficient LLM pre-training with reduced memory and computation.

Ax Wenjin Hou, Shangpin Peng, Weinong Wang, Zheng Ruan, Yue Zhang, Zhenglin Zhou, Mingqi Gao, Yifei Chen, Kaiqi Wang, Hongming Yang, Chengquan Zhang, Zhuotao Tian, Han Hu, Yi Yang, Fei Wu, Hehe Fan 5/6/2026

Uni-OPD: Unifying On-Policy Distillation with a Dual-Perspective Recipe

Uni-OPD framework unifies on-policy distillation theory, identifying bottlenecks in consolidating expert models into single student models.

Ax Mahmoud Hanouneh, Radu Timofte, Dmitry Ignatov 5/6/2026

From Code to Prediction: Fine-Tuning LLMs for Neural Network Performance Classification in NNGPT

Fine-tuned LLMs for predicting neural network performance across datasets in AutoML frameworks, bridging code generation and performance reasoning.

Ax Filip Kronstr\"om, Alexander H. Gower, Daniel Brunns{\aa}ker, Ievgeniia A. Tiukova, Ross D. King 5/6/2026

Graph Neural Network based Hierarchy-Aware Embeddings of Knowledge Graphs: Applications to Yeast Phenotype Prediction

GNN-based method for hierarchy-aware knowledge graph embeddings using semantic loss from ontologies, applied to yeast gene deletion phenotype prediction.

Ax Meng Xiang, Yan Pei 5/6/2026

Distribution-Free Pretraining of Classification Losses via Evolutionary Dynamics

Evolutionary Dynamic Loss framework learns transferable classification losses using synthetic data and ranking-consistency objectives without real samples.

Ax Juneyoung Park 5/6/2026

Rethinking the Rank Threshold for LoRA Fine-Tuning

Re-examines LoRA rank thresholds for LLM fine-tuning, questioning prior NTK analysis and its practical implications for cross-entropy loss.

Ax Jinghui Yuan, Jiaxuan Zou, Shuo Wang, Yong Liu, Feiping Nie 5/6/2026

Nora: Normalized Orthogonal Row Alignment for Scalable Matrix Optimizer

Introduces Nora, a matrix-based optimizer for LLMs combining Muon-like preconditioning with scale-invariance and computational efficiency.

Ax Ryan King, Gang Li, Bobak Mortazavi, Tianbao Yang 5/6/2026

Memory-Efficient Continual Learning with CLIP Models

Proposes memory-efficient continual learning for CLIP models using distributed contrastive loss to reduce forgetting with small memory buffers.

Ax Hassan Dbouk, Nidham Gazagnadou, Matthias Reisser, Christos Louizos 5/6/2026

On Adaptivity in Zeroth-Order Optimization

Analyzes zeroth-order optimization for LLM fine-tuning, showing adaptive ZO methods offer no convergence advantage over ZO-SGD despite memory overhead.

Ax Eszter Varga-Umbrich, Shikha Surana, Paul Duckworth, Jules Tilly, Olivier Peltre, Zachary Weller-Davies 5/6/2026

Pretrained Model Representations as Acquisition Signals for Active Learning of MLIPs

Uses pretrained MLIP representations as acquisition signals for active learning in machine learning interatomic potentials for chemistry.

Ax Hikmat Karimov, Rahid Zahid Alekberli 5/6/2026

An Information-Geometric Framework for Stability Analysis of Large Language Models under Entropic Stress

Thermodynamic-inspired framework for analyzing LLM stability under uncertainty using entropy-based metrics beyond aggregate accuracy.

Ax Hikmat Karimov, Rahid Zahid Alekberli 5/6/2026

The Kerimov-Alekberli Model: An Information-Geometric Framework for Real-Time System Stability

Kerimov-Alekberli model applying information geometry and non-equilibrium thermodynamics to AI safety and autonomous system alignment.

Ax Bumjun Kim, Albert No 5/6/2026

Memorization In Stable Diffusion Is Unexpectedly Driven by CLIP Embeddings

Analysis of how CLIP embeddings unexpectedly drive memorization in Stable Diffusion text-to-image models, with categorization of token types.

Ax Cheng Qian, Hyeonjeong Ha, Jiayu Liu, Bingxiang He, Jeonghwan Kim, Jiateng Liu, Bingxuan Li, Aditi Tiwari, Dwip Dalal, Zhenhailong Wang, Xiusi Chen, Mahdi Namazifar, Yunzhu Li, Heng Ji 5/6/2026

CreativityBench: Evaluating Agent Creative Reasoning via Affordance-Based Tool Repurposing

CreativityBench benchmark evaluating LLM creative problem-solving through affordance-based tool repurposing tasks.

Ax Sakshi Agarwal, Aishik Konwer, Ankit Parag Shah 5/6/2026

Reasoning-Guided Grounding: Elevating Video Anomaly Detection through Multimodal Large Language Models

VANGUARD system using multimodal LLMs for video anomaly detection with interpretable reasoning and spatial localization of anomalous events.

Ax Aditya Ajay Phalod 5/6/2026

When Should a Language Model Trust Itself? Same-Model Self-Verification as a Conditional Confidence Signal

Evaluation of same-model self-verification as confidence signal for LLMs compared against likelihood-based baselines on reasoning tasks.

Ax Gavin Money, Sindhuja Penchala, Jiacheng Li, Noorbakhsh Amiri Golilarz 5/6/2026

Where to Bind Matters: Hebbian Fast Weights in Vision Transformers for Few-Shot Character Recognition

Study of Hebbian Fast-Weight modules in vision transformers enabling rapid within-episode adaptation for few-shot character recognition.

Ax Rui Tang, Kaiyu Xu, Pengsen Cheng, Hao Ren, Haizhou Wang, Shuyu Jiang 5/6/2026

EvoJail: Evolutionary Diverse Jailbreak Prompt Generation for Large Language Models

EvoJail method for generating diverse automated jailbreak prompts for LLMs using evolutionary algorithms, designed for adaptability to updated safety-finetuned models.

Ax Shannon K. Gallagher, Swati Rallapalli, Tyler Brooks, Chuck Loughin, Michele Sezgin, Ronald Yurko 5/6/2026

Analysis and Explainability of LLMs Via Evolutionary Methods

Framework applying evolutionary methods to analyze and explain LLMs by relating model weights to genotypes and outputs to phenotypes.

Ax Muhammad Muneeb, David B. Ascher 5/6/2026

EFGPP: Exploratory framework for genotype-phenotype prediction

EFGPP framework for genotype-to-phenotype prediction combining multiple data sources for complex trait prediction, tested on migraine prediction.

Ax Xulin Hu, Che Wang, Wei Yang Bryan Lim, Jianbo Gao, Zhong Chen 5/6/2026

Tracing the Dynamics of Refusal: Exploiting Latent Refusal Trajectories for Robust Jailbreak Detection

SALO method for detecting jailbreaks by tracing dynamic refusal trajectories in LLM representations using causal analysis, robust against adversarial attacks.

Ax Yingyun Li, Yu Wang, Haiyang Qian 5/6/2026

MedStruct-S: A Benchmark for Key Discovery, Key-Conditioned QA and Semi-Structured Extraction from OCR Clinical Reports

MedStruct-S benchmark for semi-structured information extraction from OCR clinical reports with tasks for key discovery, QA, and key-value pair extraction.

Ax Yucheng Du 5/6/2026

Geometric Deviation as an Unsupervised Pre-Generation Reliability Signal: Probing LLM Representations for Answerability

Method to detect whether LLM can answer queries before generation by measuring hidden state geometry deviation from answerable reference sets, tested on Llama, Qwen, and Mistral models.

Ax Prakhar Gupta, Garv Shah, Satyam Goyal, Anirudh Kanchi 5/6/2026

Sparse Memory Finetuning as a Low-Forgetting Alternative to LoRA and Full Finetuning

Sparse Memory Finetuning (SMF) method for adapting LLMs while reducing catastrophic forgetting by selectively updating memory rows during training. Compared against LoRA and full finetuning.

Ax Dongyoung Kim, Huiwon Jang, Myungkyu Koo, Suhyeok Jang, Taeyoung Kim, Beomjun Kim, Byungjun Yoon, Changsung Jang, Daewon Choi, Dongsu Han, Donguk Lee, Heeseung Kwon, Hojin Jeon, Jaehyun Kang, Jaekyoung Bae, Jihyuk Lee, Jimin Lee, John Won, Joonwoo Ahn, Junhyeong Park, Junyoung Sung, Kyungmin Lee, Minseong Han, Minsung Yoon, Sejune Joo, Seonil Son, Seungcheol Park, Seunggeun Cho, Seungjun Moon, Seungku Kim, Yonghoon Dong, Yongjin Cho, Youngchan Kim, Chang Hwan Kim, Dohyeon Kim, Hazel Lee, Heecheol Kim, Hensen Ahn, Hyungkyu Ryu, Hyunsoo Choi, Hyunsoo Shin, Jaeheon Jung, Jaewoo Kim, Jinwook Kim, Joochul Chang, Joonsoo Kim, Junghun Park, Jungwoo Park, Junho Cho, Junhyeok Park, Junwon Lee, Kangwook Lee, Kwanghoon Kim, Kyoungwhan Choe, Manoj Bhadu, Nayoung Oh, Sangjun Kim, Sangwoo Kim, Seunghoon Shim, Seunghyun Kim, Seungjun Lee, Seungyup Ka, Sungryol Yang, Wook Jung, Yashu Shukla, Yeonjae Lee, Yeonwoo Bae, Jinwoo Shin 5/6/2026

RLDX-1 Technical Report

RLDX-1 robotic policy model extending Vision-Language-Action models with motion awareness, memory, and physical sensing for complex real-world tasks.

Ax Daniel Fraiman, Ricardo Fraiman 5/6/2026

Imbalanced Classification under Capacity Constraints

Imbalanced classification framework addressing underrepresented classes under operational capacity constraints for rare detection tasks.

Ax Van-Phat Thai, Aradhya Dhruv, Duc-Thinh Pham, Sameer Alam 5/6/2026

Contrastive Regularization for Accent-Robust ASR

Contrastive regularization method for accent-robust ASR using supervised contrastive learning as auxiliary CTC fine-tuning objective.

Ax Maksym Nechepurenko, Pavel Shuvalov 5/6/2026

Coordination as an Architectural Layer for LLM-Based Multi-Agent Systems

Treats coordination as architectural layer for multi-agent LLM systems, addressing 41-87% production failure rates from coordination defects not capability gaps.

Ax Dionysis Arvanitakis, Vaggos Chatziafratis, Yiyuan Luo 5/6/2026

Provable Accuracy Collapse in Embedding-Based Representations under Dimensionality Mismatch

Proves sharp dimension-accuracy tradeoffs in embedding representations showing accuracy collapse when embedding dimension mismatches ground truth.

Ax Jaavid Aktar Husain, Dorien Herremans 5/6/2026

APEX: Large-scale Multi-task Aesthetic-Informed Popularity Prediction for AI-Generated Music

APEX predicts popularity for AI-generated music using multi-task learning with aesthetic quality metrics on new AI music landscape.

Ax Yujun Li, Hongyuan Zhang, Yuan Yuan 5/6/2026

GRPO-TTA: Test-Time Visual Tuning for Vision-Language Models via GRPO-Driven Reinforcement Learning

GRPO-TTA applies Group Relative Policy Optimization to test-time adaptation of vision-language models via prompt refinement and RL.

Ax Haoyu Zhang, Mohammad Zandsalimy, Shanu Sushmita 5/6/2026

Exposing LLM Safety Gaps Through Mathematical Encoding:New Attacks and Systematic Analysis

Demonstrates LLM safety bypasses using mathematical encoding (set theory, logic, quantum mechanics) achieving 46-56% attack success across eight models.

Ax Seunghan Lee, Jun Seo, Jaehoon Lee, Sungdong Yoo, Minjae Kim, Tae Yoon Lim, Dongwan Kang, Hwanil Choi, Soonyoung Lee, Wonbin Ahn 5/6/2026

FinSTaR: Towards Financial Reasoning with Time Series Reasoning Models

Proposes capability taxonomy for time series reasoning models applied to financial domain with 2x2 classification framework for entity and temporal analysis.

Ax Ishrith Gowda (University of California, Berkeley) 5/6/2026

MEMSAD: Gradient-Coupled Anomaly Detection for Memory Poisoning in Retrieval-Augmented Agents

Formalizes memory poisoning attacks on retrieval-augmented LLM agents using game theory, proposing MEMSAD detection method with unified evaluation framework.

Ax Zhongjian Zhang, Yue Yu, Mengmei Zhang, Junping Du, Xiao Wang, Chuan Shi 5/6/2026

Revisiting Graph-Tokenizing Large Language Models: A Systematic Evaluation of Graph Token Understanding

Systematic evaluation of Graph-Tokenizing LLMs questioning effectiveness of treating graph data as prefix tokens for LLMs.

Ax Jingxi Qiu, Zeyu Han, Cheng Huang 5/6/2026

SURE-RAG: Sufficiency and Uncertainty-Aware Evidence Verification for Selective Retrieval-Augmented Generation

SURE-RAG framework for selective retrieval-augmented generation that verifies evidence sufficiency before answering.

Ax Zirui Tang, Xuanhe Zhou, Yumou Liu, Linchun Li, Weizheng Wang, Hongzhang Huang, Jun Zhou, Jiachen Song, Shaoli Yu, Jinqi Wang, Zihang Zhou, Hongyi Zhou, Yuting Lv, Jinyang Li, Jiashuo Liu, Ruoyu Chen, Chunwei Liu, GuoLiang Li, Jihua Kang, Fan Wu 5/6/2026