Hacia dónde va la investigación en IA

Las expresiones que aparecen mucho más a menudo en los artículos publicados en los últimos 30 días que de costumbre. Un buen indicio de los asuntos que despuntan.

1. ExpresiónArtículos8 sem.Evolución
2. 01▶training seeds16+124 %

   • Auditing Routing Entropy as an Uncertainty Signal in Attention-Residual Transformers2 oct · Wenhao Liang, Lin Yue, Wei Emma Zhang, Mingyu Guo, Olaf Maennel, Weitong Chen
   • Mathematical Transfer in LLMs Follows Reasoning Approach More Than Topic2 oct · Sajad Goudarzi, Samaneh Zamanifard, Seyed Amin Seyed Haeri, Moloud Nasiri, Hamed Rahimian
   • Predictive Safety Curricula for Robust Legged Locomotion30 sept · Ivan Ovinnikov, Pascal Sutter, Christian Gehring, Jordis Herrmann
   • Energy Vision--Language--Action: A Controlled Multimodal Benchmark for Intent-Conditioned Residential Energy Management29 sept · Lyes Saad Saoud, Oualid Doukhi, Ehsan Reihani, Saeed Sepasi, Deok Jin Lee, Moussa Ayyash, Reza Ghorbani
   • A Persistent State for Auditable Mixture-of-Experts Routing29 sept · Abdurrahman Javat, Allan Kazakov
   • Query-Conditioned Prototype Adaptation for Cross-Domain Few-Shot Learning: Single-Query Inference, Controlled Comparisons, and Failure Modes28 sept · Rushab Rasik Karania, Tomas Maul
   • Global tree forecasters collapse at the hierarchical aggregate: a five-panel failure characterization24 sept · Md Rezwanul Islam, Wael Mohammed
   • CleanScore: Black-Box Benchmark Audits with Negative Controls and Sensitivity Bounds22 sept · Jeffery Opoku, David Banahene
   • Brain-to-Image Generation: Reconstructing Visual Stimuli from EEG using Generative Adversarial Networks22 sept · Harshit Goyal
   • Reliability-Centered Evaluation of Sparse Longitudinal CT Lesion-Size Forecasting with Conformal Interval Calibration and Gompertz-Inspired Regularization21 sept · Lingfei Kong
   • Los 10 más recientes que citan «training seeds».
3. 02▶naturally occurring20+95 %

   • VietPrism: A large-scale Vietnamese speech and deepfake corpus with diverse dialects and code-switching25 sept · Minh Hoang, Thai Le
   • ThaiTrees: Thai Syntactic Dependency Trees Across Domains24 sept · Attapol T. Rutherford, Papatchol Thientong
   • AraGenre 2026: A Hierarchical Definition-Guided Arabic Genre Classification Shared Task24 sept · Mo El-Haj, Saad Ezzini, Shadi Abudalfa, Mustafa Jarrar, Nguyen Minh Chi, Nguyen Minh Quan
   • On the Lexical Superstition of Large Language Models for Code Comprehension: Re-evaluation on Code of Low Lexical Quality23 sept · Xin Shen (Nanjing University, Nanjing, China), San-Zhuo Xi (Nanjing University, Nanjing, China), Yali Du (Nanjing University, Nanjing, China), Ming Li (Nanjing University, Nanjing, China)
   • Anatomy of a Closed-Loop Collapse: A Causal Case Study of a Compressed VLA Policy22 sept · Fengze Jia (The Ohio State University)
   • AskQE: Question Answering as Automatic Evaluation for Machine Translation22 sept · Dayeon Ki, Kevin Duh, Marine Carpuat
   • MuLA-Bench: A Multilingual Long-Form Audio Understanding Benchmark via Multi-Tier Auditing22 sept · Zeyu Yang, Xinyu Zhang, Zibo Bi, Pei Zhang, Xize Cheng, Jin Xu, Baosong Yang, Satoshi Nakamura
   • TEAR: Table Extraction with Attribute Recommendation from Texts via Large Language Models15 sept · Tong Li, Shuye Ding, Jiachuan Wang, Yongqi Zhang, Shuangyin Li, Lei Chen, Bo Li
   • Context-Aware Causal Gaze Forecasting for Human-Vehicle Interaction During In-Cabin Tracking Dropouts14 sept · Shabnam Shabani, Ghazal Farhani
   • WinSyn: An Automated Pipeline for Realistic Enterprise Question-Answering Evaluation14 sept · Amey Varhade, Ananya Sutradhar, Ravishankar Krishnaswamy, Navin Goyal
   • Los 10 más recientes que citan «naturally occurring».
4. 03▶language-model agents26+93 %

   • Managing Context and Communication in Distributed Agentic UAV Swarms2 oct · Andrea Iannoli, Ivan Zyrianoff, Angelo Trotta, Lorenzo Gigli, Marco Di Felice
   • Harness as a Language: A Minimalist Agent Framework With Maximal Expressivity1 oct · Zhening Li, Joshua Liu, Mateja Vukelic, Nicole Shen, Supriya Lall, Amitayush Thakur, Alex Zhang, Omar Khattab, Jonathan Light, Armando Solar-Lezama
   • From Imitation to Reward Discovery: On-Policy Warmup for Agentic RL1 oct · Yitong Qiao, Tiantian He, Lei Liu, Yue Shen, Jian Wang, Jinjie Gu, Zhixuan Chu
   • OverForge: Reasoning Through Strategies and Tactics Helps Cooperative Lifelong Adaptation1 oct · Oana Madalina Fron, Ojas Shirekar, Chirag Raman
   • Coding Agent Memory Post-training: Unlocking the Memory Potential of Pre-trained File Operations for Long-Horizon Tasks via Reinforcement Learning30 sept · Lirui Luo, Kelong Mao, Heming Xia, Rongqing Li, Xinwei Yang, Luyu Chen, Kieran Wong, Yudong Guo, Xinrui Wang, Jiayin Zhu, Simiu Gu, Sulong Xu, Cong Fang
   • SINGED: Correct Outputs Do Not Certify Safe Execution in LLM Agents30 sept · Xiaoyu Xu, Zi Liang, Minxin Du, Qipeng Xie, Qingqing Ye, Yuyuan Li, Haibo Hu
   • Boundary-State Control for Tool-Using Language-Model Agents: Commit-Time Consistency under State Drift30 sept · Wesley Shu
   • Semantic Projection for Continual Self-Evolution of Language Agents30 sept · Ziyu Liu, Jun Chen, Lixu Wang
   • DGF-Bench: A Benchmark for Simulating and Auditing Deception Against Multi-Agent Governance Boards29 sept · Jeremy Canale
   • Social Circuits behind Multi-agent Echo Chambers29 sept · Chuiyang Meng, Wenlu Yu, Ming Tang, Cheng Li
   • Los 10 más recientes que citan «language-model agents».
5. 04▶operating characteristic curve18+85 %

   • Partial AUC Maximization from Positive-unlabeled Data2 oct · Atsutoshi Kumagai, Tomoharu Iwata, Taishi Nishiyama, Hiroshi Takahashi, Kazuki Adachi, Yasuhiro Fujiwara
   • Storage Is Not Strategy: State-Conditioned Support Control for LLM Unlearning1 oct · Tianhao Qian, Ziming Hong, Chongyang Gao, Kezhen Chen, Lixu Wang
   • RGDT-Bench: Benchmarking LLM Reasoning for Rule-Governed Decisions and Their Justifications30 sept · Jianpeng Zhao, Haihua Xu, Haoyang Zhang, Shuang Qian, Yixiang Tang, Xintao Wang, Kun Sun, Pei Wu, Shuhan Zhong, Pengyang Wang
   • Unsupervised spiking feature learning for event-based pedestrian crossing detection: approaching supervised accuracy without labelled training data29 sept · Henok Teklu, Mustafa Sakhai, Matej Mertik, Maciej Wielgosz
   • Different Corruptions, Different Signals: Uncertainty and Loss in Federated Data Quality28 sept · Bradley Scott, Zeqi Luo, Edmond S. L. Ho
   • Auditing Latent-Space Monitors for Autonomous Driving28 sept · Nikhil Kamalkumar Advani, Vishwajeet Shivaji Hogale, Saurav Kumar
   • Uncovering Residential PV-EV Co-Adoption from Smart-Meter Data: Load Archetypes and Detection for Demand-Side Planning25 sept · Jack Zheng, Hao Wang
   • Radiomics-Conditioned Modulation of RenalCLIP Features for Clear Cell Renal Cell Carcinoma Classification23 sept · Yuan Liang, Sourav Bhattacharjee, Abraham Campbell
   • Interpretable AI plus Handheld, Portable Retinal Photographs: A Low-Cost Glaucoma Screening Solution for West Africa23 sept · Charis Y. N. Chiang, Tarela Sarimiye, Adeyinka Ashaye, Martin Buist, Michael A. Hauser, Olusola Olawoye, Micha\"el J. A. Girard
   • Early Prediction of Pathological Complete Response to Neoadjuvant Chemotherapy Using Temporal Deep Learning on DWI23 sept · Pablo Garc\'ia Marcos, Md. Tarequl Islam, Paula Puerta Gonz\'alez, Guillermo Lorenzo, H\'ector G\'omez, Covadonga del Camino, Angel Rio-Alvarez, V\'ictor M. Gonz\'alez
   • Los 10 más recientes que citan «operating characteristic curve».
6. 05▶controlled ablations21+82 %

   • Generalist Representation, Specialist Detection: TS-Router for Time-Series Anomaly Detection2 oct · Tian Lan, Yifei Gao, Yimeng Lu, Xuming An, Meng Wang, Yue Pan, Wenjun He, Chenghao Liu, Chen Zhang
   • What Limits Recursive Reasoning Models: Optimization, Architecture and Test-Time Scaling1 oct · Yuliana Shakhvalieva, Dmitrii Kharchev, Viacheslav Bezrukov, Inessa Fedorova, Dmitry Bocharov, Ivan Oseledets, Valerii Ternovskii
   • DualGuard: Dual-Mode Quality Control for Logic-Preserving Data Augmentation30 sept · Shenghao Li, Lin Zhao
   • RareDx: Controlled Knowledge Integration and Graph-Grounded Policy Optimization for Rare-Disease Diagnosis29 sept · Bo Zhang, Yuchen Wang, Dongbai Li, Matthew Yu Heng Wong, Qingkai Zeng, Lijun Wang, Tien-Yin Wong, Peng Cui, Tianyu Liu
   • RelaxKV: Recomputation Guided by the Query with Sparse Context Attention for Efficient KV Cache Reuse29 sept · Ruoling Qi, Yirui Liu, Xuaner Wu, Yuxin Jin, Jian Chen, Jiayu Qin, Yin Chen, Jiawei Shao
   • Operator Packages, Proposer Strength, and Construction-Family Plateaus in Office-Scale Verified Search25 sept · Roberto I. Ono Filho
   • AstraLOD3: Zero-shot multimodal agentic reconstruction of LOD3 building models24 sept · Bryan G. Pantoja-Rosero
   • MultiViewDx: Evidence-Linked Multi-View Clinical Diagnosis23 sept · Junda Wang, Zonghai Yao, Yujan Ting, Eric Z. Chen, Hieu Tran, Hong Yu, Weijing Huang, Terrence Chen
   • Behavior is Not Enough: A Mechanism-Based Evaluation of Social Norm Emergence in LLM Societies23 sept · Rasika Muralidharan, Haewoon Kwak, Jisun An
   • Omni2Web: Benchmarking Audiovisual Website Development22 sept · Minghao Han, Zhenghao Xing, Xize Cheng, Yuxuan Wang, Junming Lin, Ling Wang, Yinsong Yan, Yunfei Chu, Qize Yang, Jin Xu
   • Los 10 más recientes que citan «controlled ablations».
7. 06▶surface form18+70 %

   • System Attribution in LLM Brand Recommendations: Single Responses Identify the System, Aggregated Brand Profiles Do Not Transfer2 oct · Dmitrij \.Zatuchin
   • Ontology-Grounded, Reasoner-Verified Benchmarks for Evaluating LLM Reasoning in Scientific AI2 oct · Nishtha N. Vaidya, Stephan Grimm, Thomas Hubauer, Thomas A. Runkler
   • A Character-Level Neural Approach to Sinhala Sandhi Splitting1 oct · Yasas Ekanayaka, Deshan Sumanathilaka
   • KinyaMed: Seeds, Not Rows -- What a Corpus Requirement Written in the Wrong Unit Fails to Constrain30 sept · Marius Bayizere
   • What Drives Dialectal Jailbreaks? An Ablation of Surface Form, Cultural Framing, and Strategy Banks29 sept · Qingyang Xu
   • Low-Cost Assays for Measuring Model Behavior Across Vendors and Releases25 sept · Tapan Parikh
   • How Much Were You Told? Measuring External Information in Peer Reviews24 sept · Matthieu Dubois, Pablo Piantanida, Fran\c{c}ois Yvon
   • Phonemizing User-Generated Text: A Benchmark, Taxonomy, and Compositional Approach24 sept · MinJu Jeon, Younghan Park, Han Sung Park, Jong-Hwan Kim, Dong-Jin Kim, Hoyeon Lee
   • LexLattice: Multilingual Extractive Summarization via Neural Cellular Automata on Document Hierarchies24 sept · Sujay Uday Rittikar, Sheela Ramanna
   • Testing the Construct Validity of a Functional Valence Axis in LLM Agents22 sept · Weihan Li, Xinlei Chen, Yuhan Song, Xiaofeng Lin, Tianshi Zheng
   • Los 10 más recientes que citan «surface form».
8. 07▶aggregate accuracy26+69 %

   • Generalization Is Stability, Not Accuracy: Multi-Axis Evaluation of LLMs2 oct · Nagham Omar, Mahmoud Jabarin, Maya Rozenshtein, Rom Himelstein, Avi Mendelson, Amit LeVi
   • Beyond Final Accuracy: Auditing Communication in LLM Multi-Agent Systems2 oct · Shixuan Li, Wei Yang, Peiyu Zhang, Anzhe Cheng, Heng Ping, Paul Bogdan
   • Heavy-Tailed Memory Traces in Long-Horizon Language Agents2 oct · Xinyuan Song, Zekun Cai
   • Inductive Visual Logic for Few-Shot Out-Of-Distribution Adaptation in VLMs1 oct · Hung-Jen Chen, Yu-Heng Ho, Ting-Yao Huang, Po-Hsiang Hsu, Li-Yu Chen, Chun-Yi Lee, Min Sun
   • Where Scientific Search Agents Fail: Decision-Checkpoint Auditing of Exposure and Inspection Attempts1 oct · Hongmin Li, Wanli Zhao
   • Complexity-Aware Evaluation of LLM Comprehension30 sept · Ali Mohammadi Esfahani, Nafiseh Kahani, Samuel A. Ajila
   • When Should LLMs Trust Their Own Revisions? A Risk-Aware Study of Intrinsic Self-Correction30 sept · Tianzhu Zhang
   • Pixels to Keys: Exploring Spatial and Motion Cues in Gameplay Inverse Dynamics30 sept · Abhishek Pillai, Ekta Prashnani, Joohwan Kim, Iuri Frosio
   • SemReward-VL: Semantic Reward-Guided Video-Language Adaptation for Developmental Behavior Assessment29 sept · De Jiang, Shuo Zhang, Kehong Yuan, Hongen Liao
   • Same Probe, Different Numbers: Are Activation Probes Robust to Inference-Time Numerical Non-Determinism?29 sept · Alizishaan Khatri
   • Los 10 más recientes que citan «aggregate accuracy».
9. 08▶evidence supports19+68 %

   • Storage Is Not Strategy: State-Conditioned Support Control for LLM Unlearning1 oct · Tianhao Qian, Ziming Hong, Chongyang Gao, Kezhen Chen, Lixu Wang
   • AIMS: An Agentic AI Framework for Sim-to-Real Multi-Modal ISAC1 oct · Yijie Bian, Kai Zhang, Wei Guo, Zixin Wang, Shenghui Song, Jun Zhang, Khaled B. Letaief
   • BELIEFRAG: Making Adaptive RAG State-Aware under Evolving Evidence1 oct · Hongji Pu
   • The Argument and the Letterhead: Source-Position Coherence in AI Evaluation29 sept · Michele Loi
   • DISCERN: Can AI Agents Work Like Scientists and Guide Discovery?29 sept · Nan Huang, Mario Tapia-Pacheco, Kun Zhou, Yiming Huang, Kevin Jos\'e Barrientos D\'iaz, Tiffany Amariuta, Jingbo Shang
   • How Reproducible Are Evaluation Conclusions? A Self-Audit of LLM-Inferred Prompt Structure25 sept · Dipankar Sarkar
   • Judgment-Centred Software Engineering Education: A Post-Hype Review and Framework for AI-Augmented Learning25 sept · Qusay H. Mahmoud
   • SGDet3D++: Geometry-Grounded Semantics for 4D Radar and Camera 3D Object Detection24 sept · Xiaokai Bai, Zhenyu Fan, Lianqing Zheng, Songkai Wang, Si-Yuan Cao, Hui-liang Shen
   • From Offline Proxies to Online Decisions: A Layered Engagement Evaluation Framework for Conversational AI23 sept · Xuanyi Li, Vaskar Nath, Hossein Amirkhani, Jay Li, Alex Deng
   • Active Visual Sampling with a Connectome-Constrained Fly Model for One-Shot Hatch Recognition in Architectural Drawings22 sept · Dmitry Kuklev
   • Los 10 más recientes que citan «evidence supports».
10. 09▶execution feedback31+67 %

    • Reconstruct, Practice, Go Real: Guided Self-Improvement for Embodied Agents2 oct · Yen-Jen Wang, Haozhe Jiang, Shuying Deng, Haoru Xue, Weirui Ye, Rocky Duan, Nika Haghtalab, S. Shankar Sastry, Pieter Abbeel, Haozhi Qi
    • VideoEvolve: Evolving Agent Harnesses for Video Temporal Grounding2 oct · Bingjun Luo, Yuhuan Fan, Jialin Guo, Siqi Li
    • ActiveSaddler: Automated Curriculum Learning for Agent Harness Optimization2 oct · Sungho Park, Wonjoong Kim, Jue Zhang, Wook-Shin Han, Pengfei Gao, Chanyoung Park, Yongqiang Yao, Rao Fu, Elsie Nallipogu, Qingwei Lin, Victor R\"uhle
    • VidHarness: Evolving Agent Harnesses for Cost-Efficient Long Video Understanding1 oct · Susan Liang, Jianmin Wu, Daxiang Dong
    • HeurEvo: Agentic Evolution of Hybrid Solver-Augmented Heuristics for Time-Critical Mathematical Optimization1 oct · Feijie Wu, Hugo Barbalho, Konstantina Mellou, Marco Molinaro, Jing Gao, Ishai Menache, Xinzhi Zhang, Sirui Li
    • Learning When and How to Intervene: A Hindsight-Distilled Sentinel for Coding Agents1 oct · Jiangrui Zhao, Chenglong Li, Meng Zhang, Xiaoting Du
    • Make Code as Policy Great Again: Frontier Agents Write, Call, and Evolve Robot Tools1 oct · Shijia Ge, Alex Zhou, Jianshu Zeng, Yexing Wan, Di Wu, Zelin Zheng, Yazhe Wang, Zhiqi Jia, Xuan Shangguan, Jay Zhu, Yijun Liu, Lingyu He, Sihang Wu, Xiao He, Hongcheng Gao
    • Learning from Research: Toward Lifelong Agent Harness Evolution1 oct · Jingbo Yang, Kwei-Herng Lai, Xiaowen Wang, Yaar Harari, Evgeniy Gabrilovich, Shiyu Chang
    • Harness Learning Enables Generalizable Test-Time Adaptation30 sept · Alvin Zhang, Xuecheng Liu, Zixuan Wang, Fahim Tajwar, Daman Arora, Ruslan Salakhutdinov, Daniel Khashabi, Yuda Song, Andrea Zanette
    • Learning Meta-Skills for Agent Harness Design in Test-Time AI4AI30 sept · Cheng Qian, Kunlun Zhu, Beibin Li, Zhenhailong Wang, Heng Ji
    • Los 10 más recientes que citan «execution feedback».
11. 10▶complementary roles17+67 %

    • LLM-Driven Multi-Agent Control for Skill-Based Smart Manufacturing2 oct · Kay K\"ohle, Darko Anicic, Thomas A. Runkler, Ren\'e Graf
    • LabBook: Harnessing Experimental History for Efficient LLM-Driven Discovery2 oct · Bo Yuan, Wenqian Ye, Zelin Zhao, Lama Moukheiber, Henry Kautz, Aidong Zhang, Yongxin Chen
    • Less Data, Better Timing: Student-Curriculum Coupling for VLM On-Policy Distillation in Temporal Video Grounding1 oct · Jiacheng Qiu, Yunsoo Kim, Ruichen Xu, Jian Luo, Petar M. Djuri\'c, Sima Mofakham
    • DualGuard: Dual-Mode Quality Control for Logic-Preserving Data Augmentation30 sept · Shenghao Li, Lin Zhao
    • EHRAdapt: Adapting Pretrained Language Models to Electronic Health Records with Semantic Priors for Rare Clinical Events29 sept · Andre R Goncalves, Vincent Liu, Priyadip Ray
    • Semantic Navigation for Issue Localization in Code Repository28 sept · Yunxiang Wei, Zhenyu Lei, Jundong Li
    • From Alignment to Fusion in 3D Vision-Language24 sept · Xueqi Qiu, Xingyu Miao, Jingjing Deng, Haoran Duan, Yang Long, Ling Shao
    • Complementary Roles of Activation and Parametric Memory in Few-Shot Learning24 sept · Miaohe Niu, Runsong Zhao, Xinyu Liu, Bo Jin, Yucheng Qiao, Chunliang Zhang, Jingbo Zhu, Tong Xiao
    • Robust Photoplethysmography Signal Denoising via Mamba Networks23 sept · I Chiu, Yu-Tung Liu, Kuan-Chen Wang, Hung-Yu Wei, Yu Tsao
    • Complementary Roles of Radiomics and Foundation Representations in Renal Cell Carcinoma Classification: A Comparative Study of 2D and 3D CT Encodings23 sept · Yuan Liang, Sourav Bhattacharjee, Abraham Campbell
    • Los 10 más recientes que citan «complementary roles».

Haz clic en una expresión para ver los artículos que la emplean. Las expresiones se extraen automáticamente del título y del resumen: son grupos de 2 o 3 palabras, en inglés y en minúsculas, tal como aparecen en el texto original. La evolución compara los últimos 30 días con los 6 meses anteriores.

Asuntos procedentes de la clasificación científica de OpenAlex.

Explorar los temas →