arXiv cs.AI Analysis Report - 20251003

Analysis of arXiv cs.AI Papers

📊 数据统计概览

📈基本统计

  • 论文总数: 100
  • 分析分类: cs.AI
  • 时间范围: 20251003
  • 独立作者数: 536

👥高产作者 Top 10

  1. Yun Fu (2 篇)
  2. Jian Zhang (2 篇)
  3. Bo Ma (2 篇)
  4. Hang Li (2 篇)
  5. ZeHua Hu (2 篇)
  6. XiaoFan Gui (2 篇)
  7. LuYao Liu (2 篇)
  8. Simon Liu (2 篇)
  9. Tianyi Zhang (2 篇)
  10. Shivali Dalmia (2 篇)

🔍热门关键词 Top 10

  1. language (51 次)
  2. learning (41 次)
  3. reasoning (36 次)
  4. llms (33 次)
  5. such (26 次)
  6. data (26 次)
  7. across (22 次)
  8. generation (22 次)
  9. diffusion (20 次)
  10. llm (19 次)

Publication Date: 2025-10-03 | Papers Analyzed: 100

1. Research Hot Topics Analysis

The analysis of 100 papers from arXiv's cs.AI category reveals several dominant research trends. The following topics represent the most active areas of investigation, with a significant focus on enhancing the capabilities, safety, and application of Large Language Models (LLMs).

Large Language Models (LLMs)

Core Focus: The undeniable center of gravity. Research spans enhancing LLM reasoning, developing agentic frameworks (e.g., for data visualization, program repair), creating evaluation benchmarks (TutorBench), and improving alignment. There is a strong push towards making LLMs more autonomous and reliable in complex, multi-step tasks.

AI Safety, Robustness & Alignment

Core Focus: A critical and growing area. This includes developing methods for "jailbreak" detection in multi-turn dialogues (NEXUS), creating policy-compliant agents (PolicyGuardBench), red-teaming multimodal models (ARMs), and ensuring robustness against adversarial attacks. The goal is to build safer and more controllable AI systems.

Multimodality & Vision-Language Models (VLMs)

Core Focus: Integrating language with other data types, primarily vision. Key efforts include universal medical image segmentation (DuPLUS), virtual try-on technologies (DiT-VTON), reconstructing visual information from brain activity (HAVIR), and improving GUI grounding for autonomous agents.

Generative AI & Diffusion Models

Core Focus: Moving beyond static generation. Research focuses on fine-tuning diffusion models with rewards, enabling concurrent text-image synthesis (OneFlow), and applying these models to specialized domains like text-to-speech (Flamed-TTS) and scientific data super-resolution.

Multi-Agent Systems

Core Focus: Designing systems where multiple AI agents collaborate. This includes frameworks for automated software engineering (ALMAS), dynamic agent architecture search (AutoMaAS), and using agents for complex tasks like data visualization (CoDA) and legal contradiction detection (LegalWiz).

Explainability & Interpretability (XAI)

Core Focus: Understanding the "black box." Research aims to disentangle recall from reasoning in transformers, analyze what RNA language models learn (SAE-RNA), and provide counterfactual explanations for smart environments, thereby increasing trust and transparency.

AI for Science & Specialized Domains

Core Focus: Applying AI to solve real-world scientific and industrial problems. This includes diagnosing schizophrenia from fMRI (BrainIB++), assessing conservation status (IUCN Red List), predicting shaft power in maritime transport, and optimizing Directed Self-Assembly (DSA) lithography.

Reinforcement Learning (RL)

Core Focus: Refining decision-making processes. RL is being used for aligning LLMs with human feedback (RLHF), developing risk-aware offline RL for safety-critical domains (RAMAC), and optimizing discrete diffusion models (MaskGRPO).

2. Author & Collaboration Analysis

The dataset features a diverse set of authors. While many papers are from distinct research groups, several clusters of collaboration are visible, indicating focused research efforts within specific teams.

Notable Author Groups

  • Bo Ma, Hang Li, et al. (Agentic AI Frameworks)
  • Keshav Ramani, Vali Tawosi, et al. (LLM Planning & Verification)
  • Shivali Dalmia, Nand Dave, et al. (Responsible & Legal AI)
  • Ashrafun Zannat, Waqas Ishtiaq, et al. (Intrusion Detection)
  • Jian Zhang, Qing Huang, et al. (Image Forgery Detection)
  • Daniel Takabi, Javad Rafiei Asl, et al. (AI Safety & Jailbreaks)

Author Collaboration Network

graph TD; subgraph Prominent Collaboration Clusters; direction LR; subgraph Agentic AI Group; Bo_Ma["Bo Ma"]; Hang_Li["Hang Li"]; ZeHua_Hu["ZeHua Hu"]; XiaoFan_Gui["XiaoFan Gui"]; Bo_Ma --- Hang_Li; Bo_Ma --- ZeHua_Hu; Hang_Li --- XiaoFan_Gui; end subgraph LLM Planning Group; Keshav_Ramani["Keshav Ramani"]; Vali_Tawosi["Vali Tawosi"]; Salwa_Alamir["Salwa Alamir"]; Keshav_Ramani --- Vali_Tawosi; Vali_Tawosi --- Salwa_Alamir; end subgraph Responsible AI Group; Shivali_Dalmia["Shivali Dalmia"]; Nand_Dave["Nand Dave"]; Abhishek_Mukherji["Abhishek Mukherji"]; Ananya_Mantravadi["Ananya Mantravadi"]; Shivali_Dalmia --- Nand_Dave; Shivali_Dalmia --- Abhishek_Mukherji; Ananya_Mantravadi --- Shivali_Dalmia; end subgraph Security & Networks Group; Ashrafun_Zannat["Ashrafun Zannat"]; Waqas_Ishtiaq["Waqas Ishtiaq"]; Md_Alamgir_Hossain["Md. Alamgir Hossain"]; Ashrafun_Zannat --- Waqas_Ishtiaq; Waqas_Ishtiaq --- Md_Alamgir_Hossain; end end

Mermaid diagram showing key collaboration clusters.

3. Technical Innovation Summary

This collection of papers introduces a wealth of novel frameworks, algorithms, and benchmarks designed to push the boundaries of AI.

Key Technology Frameworks & Paradigms

  • AgenticRAG: A zero-shot explainable recommender system using tool-augmented foundation models.
  • AutoMaAS: A self-evolving framework that uses neural architecture search to design multi-agent systems.
  • NEXUS: A framework for systematically finding and exploiting vulnerabilities in multi-turn LLM conversations.
  • UniShield: An adaptive multi-agent framework for unified forgery image detection and localization, aiming to combat misinformation.
  • OneFlow: The first non-autoregressive multimodal model enabling concurrent and mixed-modal generation of text and images.

Novel Algorithms & Methodologies

New Benchmarks & Datasets

  • TutorBench: A benchmark curated by experts to assess the tutoring capabilities of LLMs.
  • SpineBench: A clinically-grounded, level-aware benchmark for spine disorder diagnosis, powered by a 450k-sample corpus.
  • SEER: A benchmark for Span-based Emotion Evidence Retrieval, testing an LLM's ability to pinpoint exact phrases that convey emotion.

Full Paper List (100 Papers)

Title Authors
A Concept of Possibility for Real-World EventsDaniel G. Schwartz
Automated Constraint Specification for Job Scheduling by Regulating Generative Model with Domain-Specific RepresentationYu-Zhe Shi, Qiao Xu, Yanjia Li, Mingchen Liu, Huamin Qu, Lecheng Ruan, Qining Wang
Take Goodhart Seriously: Principled Limit on General-Purpose AI OptimizationAntoine Maier, Aude Maier, Tom David
Reward Model Routing in AlignmentXinle Wu, Yao Lu
Consolidating Reinforcement Learning for Multimodal Discrete Diffusion ModelsTianren Ma, Mu Zhang, Yibing Wang, Qixiang Ye
Onto-Epistemological Analysis of AI ExplanationsMartina Mattioli, Eike Petersen, Aasa Feragen, Marcello Pelillo, Siavash A. Bigdeli
A Study of Rule Omission in Raven's Progressive MatricesBinze Li
CoDA: Agentic Systems for Collaborative Data VisualizationZichen Chen, Jiefeng Chen, Sercan Ö. Arik, Misha Sra, Tomas Pfister, Jinsung Yoon
Refined Iterated Pareto Greedy for Energy-aware Hybrid Flowshop Scheduling with Blocking ConstraintsAhmed Missaoui, Cemalettin Ozturk, Barry O'Sullivan
A Qualitative Comparative Evaluation of Cognitive and Generative TheoriesPaul S. Rosenbloom
Towards Policy-Compliant Agents: Learning Efficient Guardrails For Policy Violation DetectionXiaofei Wen, Wenjie Jacky Mo, Yanan Xie, Peng Qi, Muhao Chen
OneFlow: Concurrent Mixed-Modal and Interleaved Generation with Edit FlowsJohn Nguyen, Marton Havasi, Tariq Berrada, Luke Zettlemoyer, Ricky T. Q. Chen
A Trajectory Generator for High-Density Traffic and Diverse Agent-Interaction ScenariosRuining Yang, Yi Xu, Yixiao Chen, Yun Fu, Lili Su
Automatic Building Code Review: A Case StudyHanlong Wan, Weili Xu, Michael Rosenberg, Jian Zhang, Aysha Siddika
Geolog-IA: Conversational System for Academic ThesesMicaela Fuel Pozo, Andrea Guatumillo Saltos, Yeseña Tipan Llumiquinga, Kelly Lascano Aguirre, Marilyn Castillo Jara, Christian Mejia-Escobar
When Researchers Say Mental Model/Theory of Mind of AI, What Are They Really Talking About?Xiaoyun Yin, Elmira Zahmat Doost, Shiwen Zhou, Garima Arya Yadav, Jamie C. Gorman
TutorBench: A Benchmark To Assess Tutoring Capabilities Of Large Language ModelsRakshith S Srinivasa, Zora Che, Chen Bo Calvin Zhang, Diego Mares, Ernesto Hernandez, Jayeon Park, Dean Lee, Guillermo Mangialardi, Charmaine Ng, Ed-Yeremai Hernandez Cardona, Anisha Gunjal, Yunzhong He, Bing Liu, Chen Xing
AgenticRAG: Tool-Augmented Foundation Models for Zero-Shot Explainable Recommender SystemsBo Ma, Hang Li, ZeHua Hu, XiaoFan Gui, LuYao Liu, Simon Liu
HALO: Memory-Centric Heterogeneous Accelerator with 2.5D Integration for Low-Batch LLM InferenceShubham Negi, Kaushik Roy
To Compress or Not? Pushing the Frontier of Lossless GenAI Model Weights Compression with Exponent ConcentrationZeyu Yang, Tianyi Zhang, Jianwen Xie, Chuan Li, Zhaozhuo Xu, Anshumali Shrivastava
ARMs: Adaptive Red-Teaming Agent against Multimodal Models with Plug-and-Play AttacksZhaorun Chen, Xun Liu, Mintong Kang, Jiawei Zhang, Minzhou Pan, Shuang Yang, Bo Li
Can Data-Driven Dynamics Reveal Hidden Physics? There Is A Need for Interpretable Neural OperatorsWenhan Gao, Jian Luo, Fang Wan, Ruichen Xu, Xiang Liu, Haipeng Xing, Yi Liu
Fine-Tuning Diffusion Models via Intermediate Distribution ShapingGautham Govind Anil, Shaan Ul Haque, Nithish Kannen, Dheeraj Nagaraj, Sanjay Shakkottai, Karthikeyan Shanmugam
RAMAC: Multimodal Risk-Aware Offline Reinforcement Learning and the Role of Behavior RegularizationKai Fukazawa, Kunal Mundada, Iman Soltani
A $1000\times$ Faster LLM-enhanced Algorithm For Path Planning in Large-scale Grid MapsJunlin Zeng, Xin Zhang, Xiang Zhao, Yan Pan
TravelBench : Exploring LLM Performance in Low-Resource DomainsSrinivas Billa, Xiaonan Jing
Prototyping Digital Social Spaces through Metaphor-Driven Design: Translating Spatial Concepts into an Interactive Social SimulationYoojin Hong, Martina Di Paola, Braahmi Padmakumar, Hwi Joon Lee, Mahnoor Shafiq, Joseph Seering
Fusing Multi- and Hyperspectral Satellite Data for Harmful Algal Bloom Monitoring with Self-Supervised and Hierarchical Deep LearningNicholas LaHaye, Kelly M. Luis, Michelle M. Gierach
OptunaHub: A Platform for Black-Box OptimizationYoshihiko Ozaki, Shuhei Watanabe, Toshihiko Yanase
Relevance-Aware Thresholding in Online Conformal Prediction for Time SeriesThéo Dupuy, Binbin Xu, Stéphane Perrey, Jacky Montmain, Abdelhak Imoussaten
NCV: A Node-Wise Consistency Verification Approach for Low-Cost Structured Error Localization in LLM ReasoningYulong Zhang, Li Wang, Wei Du, Peilin Li, Yuqin Dai Zhiyuan Zhao, Lingyong Fang, Ziniu Liu, Ru Zhang, Huijia Zhu, Gongshen Liu
Evaluating Large Language Models for IUCN Red List Species InformationShinya Uryu
Beyond the Final Answer: Evaluating the Reasoning Trajectories of Tool-Augmented AgentsWonjoong Kim, Sangwu Park, Yeonjun In, Sein Kim, Dongha Lee, Chanyoung Park
Knowledge-Aware Modeling with Frequency Adaptive Learning for Battery Health PrognosticsVijay Babu Pamshetti, Wei Zhang, Sumei Sun, Jie Zhang, Yonggang Wen, Qingyu Yan
Flamed-TTS: Flow Matching Attention-Free Models for Efficient Generating and Dynamic Pacing Zero-shot Text-to-SpeechHieu-Nghia Huynh-Nguyen, Huynh Nguyen Dang, Ngoc-Son Nguyen, Van Nguyen
Constraint Satisfaction Approaches to Wordle: Novel Heuristics and Cross-Lexicon ValidationJahidul Arafat, Fariha Tasmin, Sanjaya Poudel
FinReflectKG -- MultiHop: Financial QA Benchmark for Reasoning with Knowledge Graph EvidenceAbhinav Arun, Reetu Raj Harsh, Bhaskarjit Sarmah, Stefano Pasquali
FeDABoost: Fairness Aware Federated Learning with Adaptive BoostingTharuka Kasthuri Arachchige, Veselka Boeva, Shahrooz Abghari
Multimodal Carotid Risk Stratification with Large Vision-Language Models: Benchmarking, Fine-Tuning, and Clinical InsightsDaphne Tsolissou, Theofanis Ganitidis, Konstantinos Mitsis, Stergios CHristodoulidis, Maria Vakalopoulou, Konstantina Nikita
Ergodic Risk Measures: Towards a Risk-Aware Foundation for Continual Reinforcement LearningJuan Sebastian Rojas, Chi-Guhn Lee
Untargeted Jailbreak AttackXinzhe Huang, Wenjing Hu, Tianhang Zheng, Kedong Xiu, Xiaojun Jia, Di Wang, Zhan Qin, Kui Ren
From high-frequency sensors to noon reports: Using transfer learning for shaft power prediction in maritimeAkriti Sharma, Dogan Altan, Dusica Marijan, Arnbjørn Maressa
BrainIB++: Leveraging Graph Neural Networks and Information Bottleneck for Functional Brain Biomarkers in SchizophreniaTianzheng Hu, Qiang Li, Shu Liu, Vince D. Calhoun, Guido van Wingen, Shujian Yu
Learning Robust Diffusion Models from Imprecise SupervisionDong-Dong Wu, Jiacheng Cui, Wei Wang, Zhiqiang Shen, Masashi Sugiyama
Investigating The Smells of LLM Generated CodeDebalina Ghosh Paul, Hong Zhu, Ian Bayley
When and Where do Events Switch in Multi-Event Video Generation?Ruotong Liao, Guowen Huang, Qing Cheng, Thomas Seidl, Daniel Cremers, Volker Tresp
ZeroShotOpt: Towards Zero-Shot Pretrained Models for Efficient Black-Box OptimizationJamison Meindl, Yunsheng Tian, Tony Cui, Veronika Thost, Zhang-Wei Hong, Johannes Dürholt, Jie Chen, Wojciech Matusik, Mina Konaković Luković
Semantic Differentiation in Speech Emotion Recognition: Insights from Descriptive and Expressive Speech RolesRongchen Guo, Vincent Francoeur, Isar Nejadgholi, Sylvain Gagnon, Miodrag Bolic
Comparative Analysis of Parameterized Action Actor-Critic Reinforcement Learning Algorithms for Web Search Match Plan GenerationUbayd Bapoo, Clement N Nyirenda
A Unified Deep Reinforcement Learning Approach for Close Enough Traveling Salesman ProblemMingfeng Fan, Jiaqi Cheng, Yaoxin Wu, Yifeng Zhang, Yibin Yang, Guohua Wu, Guillaume Sartoretti
A Study of Neural Polar Decoders for CommunicationRom Hirsch, Ziv Aharoni, Henry D. Pfister, Haim H. Permuter
From Facts to Foils: Designing and Evaluating Counterfactual Explanations for Smart EnvironmentsAnna Trapp, Mersedeh Sadeghi, Andreas Vogelsang
HAVIR: HierArchical Vision to Image Reconstruction using CLIP-Guided Versatile DiffusionShiyi Zhang, Dong Liang, Hairong Zheng, Yihang Zhou
SpineBench: A Clinically Salient, Level-Aware Benchmark Powered by the SpineMed-450k CorpusMing Zhao, Wenhui Dong, Yang Zhang, Xiang Zheng, Zhonghao Zhang, Zian Zhou, Yunzhi Guan, Liukun Xu, Wei Peng, Zhaoyang Gong, Zhicheng Zhang, Dachuan Li, Xiaosheng Ma, Yuli Ma, Jianing Ni, Changjiang Jiang, Lixia Tian, Qixin Chen, Kaishun Xia, Pingping Liu, Tongshun Zhang, Zhiqiang Liu, Zhongan Bi, Chenyang Si, Tiansheng Sun, Caifeng Shan
UniShield: An Adaptive Multi-Agent Framework for Unified Forgery Image Detection and LocalizationQing Huang, Zhipei Xu, Xuanyu Zhang, Jian Zhang
Topic Modeling as Long-Form Generation: Can Long-Context LLMs revolutionize NTM via Zero-Shot Prompting?Xuan Xu, Haolun Li, Zhongliang Yang, Beilin Chu, Jia Song, Moxuan Xu, Linna Zhou
Coevolutionary Continuous Discrete Diffusion: Make Your Diffusion Language Model a Latent ReasonerCai Zhou, Chenxiao Yang, Yi Hu, Chenyu Wang, Chubin Zhang, Muhan Zhang, Lester Mackey, Tommi Jaakkola, Stephen Bates, Dinghuai Zhang
Abstain and Validate: A Dual-LLM Policy for Reducing Noise in Agentic Program RepairJosé Cambronero, Michele Tufano, Sherry Shi, Renyao Wei, Grant Uy, Runxiang Cheng, Chin-Jung Liu, Shiying Pan, Satish Chandra, Pat Rondon
Self-Anchor: Large Language Model Reasoning via Step-by-step Attention AlignmentHongxiang Zhang, Yuan Tian, Tianyi Zhang
Improving GUI Grounding with Explicit Position-to-Coordinate MappingSuyuchen Wang, Tianyu Zhang, Ahmed Masry, Christopher Pal, Spandana Gella, Bang Liu, Perouz Taslakian
Reward Models are Metrics in a Trench CoatSebastian Gehrmann
Diffusion-Based, Data-Assimilation-Enabled Super-Resolution of Hub-height WindsXiaolong Ma, Xu Dong, Ashley Tarrant, Lei Yang, Rao Kotamarthi, Jiali Wang, Feng Yan, Rajkumar Kettimuthu
Disentangling Recall and Reasoning in Transformer Models through Layer-wise Attention and Activation AnalysisHarshwardhan Fartale, Ashish Kattamuri, Rahul Raja, Arpita Vats, Ishita Prasad, Akshata Kishore Moharir
An Adaptive Responsible AI Governance Framework for Decentralized OrganizationsKiana Jafari Meimandi, Anka Reuel, Gabriela Aranguiz-Dias, Hatim Rahama, Ala-Eddine Ayadi, Xavier Boullier, Jérémy Verdo, Louis Montanie, Mykel Kochenderfer
TriQuest:An AI Copilot-Powered Platform for Interdisciplinary Curriculum DesignHuazhen Wang, Huimin Yang, Hainbin Lin, Yan Dong, Lili Chen, Liangliang Xia, Wenwen Xu
Can an AI-Powered Presentation Platform Based On The Game "Just a Minute" Be Used To Improve Students' Public Speaking Skills?Frederic Higham, Tommy Yuan
A Robust Clustered Federated Learning Approach for Non-IID Data with Quantity SkewMichael Ben Ali, Imen Megdiche, André Peninou, Olivier Teste
Cross-Modal Reconstruction Pretraining for Ramp Flow Prediction at Highway InterchangesYongchao Li, Jun Chen, Zhuoxuan Li, Chao Gao, Yang Li, Chu Zhang, Changyin Dong
Implicit Values Embedded in How Humans and LLMs Complete Subjective Everyday TasksArjun Arunasalam, Madison Pickering, Z. Berkay Celik, Blase Ur
Report of the 2025 Workshop on Next-Generation Ecosystems for Scientific Computing: Harnessing Community, Software, and AI for Cross-Disciplinary Team ScienceLois Curfman McInnes, Dorian Arnold, Prasanna Balaprakash, Mike Bernhardt, Beth Cerny, Anshu Dubey, Roscoe Giles, Denice Ward Hood, Mary Ann Leung, Vanessa Lopez-Marrero, Paul Messina, Olivia B. Newton, Chris Oehmen, Stefan M. Wild, Jim Willenbring, Lou Woodley, Tony Baylis, David E. Bernholdt, Chris Camano, Johannah Cohoon, Charles Ferenbaugh, Stephen M. Fiore, Sandra Gesing, Diego Gomez-Zara, James Howison, Tanzima Islam, David Kepczynski, Charles Lively, Harshitha Menon, Bronson Messer, Marieme Ngom, Umesh Paliath, Michael E. Papka, Irene Qualters, Elaine M. Raybourn, Katherine Riley, Paulina Rodriguez, Damian Rouson, Michelle Schwalbe, Sudip K. Seal, Ozge Surer, Valerie Taylor, Lingfei Wu
NEXUS: Network Exploration for eXploiting Unsafe Sequences in Multi-Turn LLM JailbreaksJavad Rafiei Asl, Sidhant Narula, Mohammad Ghasemigol, Eduardo Blanco, Daniel Takabi
LegalWiz: A Multi-Agent Generation Framework for Contradiction Detection in Legal DocumentsAnanya Mantravadi, Shivali Dalmia, Olga Pospelova, Abhishek Mukherji, Nand Dave, Anudha Mittal
ALMAS: an Autonomous LLM-based Multi-Agent Software Engineering FrameworkVali Tawosi, Keshav Ramani, Salwa Alamir, Xiaomo Liu
Bridging LLM Planning Agents and Formal Methods: A Case Study in Plan VerificationKeshav Ramani, Vali Tawosi, Salwa Alamir, Daniel Borrajo
DuPLUS: Dual-Prompt Vision-Language Framework for Universal Medical Image Segmentation and PrognosisNuman Saeed, Tausifa Jan Saleem, Fadillah Maani, Muhammad Ridzuan, Hu Wang, Mohammad Yaqub
Reasoning-based Anomaly Detection Framework: A Real-time, Scalable, and Automated Approach to Anomaly Detection Across DomainsAnupam Panwar, Himadri Pal, Jiali Chen, Kyle Cho, Riddick Jiang, Miao Zhao, Rajiv Krishnamurthy
SEER: The Span-based Emotion Evidence Retrieval BenchmarkAneesha Sampath, Oya Aran, Emily Mower Provost
AgentHub: A Research Agenda for Agent Sharing InfrastructureErik Pautsch, Tanmay Singla, Wenxin Jiang, Huiyun Peng, Behnaz Hassanshahi, Konstantin Läufer, George K. Thiruvathukal, James C. Davis
TS-Reasoner: Aligning Time Series Foundation Models with LLM ReasoningFangxu Yu, Hongyu Zhao, Tianyi Zhou
Identifying Financial Risk Information Using RAG with a Contrastive InsightAli Elahi
Triplet-Structured Knowledge Integration for Multi-Turn Medical ReasoningZhaohan Meng, Zaiqiao Meng, Siwei Liu, Iadh Ounis
Unmasking Puppeteers: Leveraging Biometric Leakage to Disarm Impersonation in AI-based VideoconferencingDanial Samadi Vahdati, Tai Duc Nguyen, Ekta Prashnani, Koki Nagano, David Luebke, Orazio Gallo, Matthew Stamm
GAS-MIL: Group-Aggregative Selection Multi-Instance Learning for Ensemble of Foundation Models in Digital Pathology Image AnalysisPeiran Quan, Zifan Gu, Zhuo Zhao, Qin Zhou, Donghan M. Yang, Ruichen Rong, Yang Xie, Guanghua Xiao
Evaluating OCR performance on food packaging labels in South AfricaMayimunah Nagayi, Alice Khan, Tamryn Frank, Rina Swart, Clement Nyirenda
DiT-VTON: Diffusion Transformer Framework for Unified Multi-Category Virtual Try-On and Virtual Try-All with Integrated Image EditingQi Li, Shuwen Qiu, Julien Han, Xingzi Xu, Mehmet Saygin Seyfioglu, Kee Kiat Koo, Karim Bouyarmane
Generative Inverse Design: From Single Point Optimization to a Diverse Design Portfolio via Conditional Variational AutoencodersMuhammad Arif Hakimi Zamrai
Transparent Reference-free Automated Evaluation of Open-Ended User Survey ResponsesSubin An, Yugyeong Ji, Junyoung Kim, Heejin Kook, Yang Lu, Josh Seltzer
CoT Referring: Improving Referring Expression Tasks with Grounded ReasoningQihua Dong, Luis Figueroa, Handong Zhao, Kushal Kafle, Jason Kuen, Zhihong Ding, Scott Cohen, Yun Fu
DynBenchmark: Customizable Ground Truths to Benchmark Community Detection and Tracking in Temporal NetworksLaurent Brisson, Cécile Bothorel, Nicolas Duminy
TRepLiNa: Layer-wise CKA+REPINA Alignment Improves Low-Resource Machine Translation in Aya-23 8BToshiki Nakai, Ravi Kiran Chikkala, Lena Sophie Oberkircher, Nicholas Jennings, Natalia Skachkova, Tatiana Anikina, Jesujoba Oluwadara Alabi
Scalable multilingual PII annotation for responsible AI in LLMsBharti Meena, Joanna Skubisz, Harshit Rajgarhia, Nand Dave, Kiran Ganesh, Shivali Dalmia, Abhishek Mukherji, Vasudevan Sundarababu
Truth-Aware Decoding: A Program-Logic Approach to Factual Language GenerationFaruk Alpay, Hamdi Alakkad
Beyond CNNs: Efficient Fine-Tuning of Multi-Modal LLMs for Object Detection on Low-Data RegimesNirmal Elamon, Rouzbeh Davoudi
Brain-Language Model Alignment: Insights into the Platonic Hypothesis and Intermediate-Layer AdvantageÁngela López-Cardona, Sebastián Idesis, Mireia Masias-Bruns, Sergi Abadal, Ioannis Arapakis
AutoMaAS: Self-Evolving Multi-Agent Architecture Search for Large Language ModelsBo Ma, Hang Li, ZeHua Hu, XiaoFan Gui, LuYao Liu, Simon Liu
A Novel Unified Lightweight Temporal-Spatial Transformer Approach for Intrusion Detection in Drone NetworksTarun Kumar Biswas, Ashrafun Zannat, Waqas Ishtiaq, Md. Alamgir Hossain
Time-To-Inconsistency: A Survival Analysis of Large Language Model Robustness to Adversarial AttacksYubo Li, Ramayya Krishnan, Rema Padman
Fully automated inverse co-optimization of templates and block copolymer blending recipes for DSA lithographyYuhao Zhou, Huangyan Shen, Qingliang Song, Qingshu Dong, Jianfeng Li, Weihua Li
CST-AFNet: A dual attention-based deep learning framework for intrusion detection in IoT networksWaqas Ishtiaq, Ashrafun Zannat, A. H. M. Shahariar Parvez, Md. Alamgir Hossain, Muntasir Hasan Kanchan, Muhammad Masud Tarek
SAE-RNA: A Sparse Autoencoder Model for Interpreting RNA Language Model RepresentationsTaehan Kim, Sangdae Nam