Trending research and the full catalog - each paper linked to the benchmarks, methods, and models it introduces.
Filtering here covers the 2,000 most recent papers, as much as one page can hold in memory. See the full index of 22,177 papers.
1 Mar 2026
Agentic vision-language models increasingly act through extended interactions, but most evaluations still focus on single-image, single-turn correctness. We introduce \textbf{AMIGO} (\textbf{A}gentic \textbf{M}ulti-\textbf{I}mage \textbf{G}rounding \textbf{O}racle Benchmark), a…
1 Mar 2026
LLM agents rerun full reasoning for every task, even one they solved moments earlier. We introduce \textbf{APEX-EM}, a non-parametric experience memory that stores complete procedural-episodic traces in a typed Procedural Knowledge Graph (PKG) and retrieves them through three…
27 Feb 2026
Large language model (LLM) agents, such as OpenAI's Operator and Claude's Computer Use, can automate workflows but unable to handle payment tasks. Existing agentic solutions have gained significant attention; however, even the latest approaches face challenges in implementing…
Dahua Lin, Xin Wang, Shiqi He · 27 Feb 2026
AIDABench is a comprehensive benchmark evaluating AI systems on complex document analysis tasks that span multiple capability dimensions and reflect real-world analytical demands across diverse industries.
Danyang Zhang, Dahua Lin, Ziqiao Ma · 23 Feb 2026
Rapid progress in video models has largely focused on visual quality, leaving their reasoning capabilities underexplored. Video reasoning grounds intelligence in spatiotemporally consistent visual environments that go beyond what text can naturally capture, enabling intuitive…
Yousof Gheisari, Mohammadreza Ghaffarzadeh-Esfahani · 21 Feb 2026
AAVGen is a generative AI framework that designs AAV capsids with improved traits through protein language models, supervised fine-tuning, and reinforcement learning techniques.
Gabriel Mongaras, Eric C. Larson · 19 Feb 2026
Researchers enhance linear attention by simplifying Mamba-2 and improving its architectural components to achieve near-softmax accuracy while maintaining memory efficiency for long sequences.
Johannes Kirmayr, Lukas Stappen, Elisabeth André · 17 Feb 2026
Users prefer adaptive feedback mechanisms in in-car AI assistants, starting with high transparency to build trust and then reducing verbosity as reliability increases, particularly in attention-critical driving scenarios.
Nihal V. Nayak, Sara Beery, David Alvarez-Melis · 16 Feb 2026
Instruction fine-tuning of large language models (LLMs) often involves selecting a subset of instruction training data from a large candidate pool, using a small query set from the target task.
Dongrui Liu, Xia Hu, Jingyi Yu · 16 Feb 2026
Clawdbot, a self-hosted AI agent with diverse tool capabilities, exhibits varying safety performance across different risk dimensions, particularly struggling with ambiguous or adversarial inputs despite consistent reliability in specified tasks.
15 Feb 2026
Predicting links in sparse, continuously evolving networks is a central challenge in network science. Conventional heuristic methods and deep learning models, including Graph Neural Networks (GNNs), are typically designed for static graphs and thus struggle to capture temporal…
Matteo Fasulo, Luca Tedeschini · 13 Feb 2026
A hierarchical approach for detecting reclaimed slurs combines weakly supervised LLM annotation with BERT-like models to incorporate LGBTQ+ community identity and sociolinguistic context into hate speech detection.
Xin Liu, Ye He, Jiawei Han · 12 Feb 2026
Embodied navigation has long been fragmented by task-specific architectures.
11 Feb 2026
Many real-world datasets contain hidden structure that cannot be detected by simple linear correlations between input features. For example, latent factors may influence the data in a coordinated way, even though their effect is invisible to covariance-based methods such as PCA.
Mu Xu, Feng Xiong, Shuang Zeng · 11 Feb 2026
Building general-purpose embodied agents across diverse hardware remains a central challenge in robotics, often framed as the ''one-brain, many-forms'' paradigm.
Shangchen Zhou, Chen Change Loy, Yihang Luo · 10 Feb 2026
4RC presents a unified feed-forward framework for 4D reconstruction from monocular videos that learns holistic scene geometry and motion dynamics through a transformer-based encoder-decoder architecture with conditional querying capabilities.
9 Feb 2026
The growing volume of video-based news content has heightened the need for transparent and reliable methods to extract on-screen information. Yet the variability of graphical layouts, typographic conventions, and platform-specific design patterns renders manual indexing…
Jakob Foerster, Tatiana Shavrina, Amar Budhiraja · 6 Feb 2026
AIRS-Bench presents a comprehensive benchmark suite for evaluating LLM agents across diverse scientific domains, demonstrating current limitations while providing open-source resources for advancement.
Arman Cohan, Yilun Zhao, Jinbiao Wei · 6 Feb 2026
A trajectory expansion framework called Anchor bootstraps scalable desktop supervision from seed demonstrations by identifying branch points and generating new trajectories through state-grounded task variants.
Zhendong Mao, Benfeng Xu, Mingxuan Du · 3 Feb 2026
Frontier language models have demonstrated strong reasoning and long-horizon tool-use capabilities.
Jinyu Xiang, Jiayi Zhang, Zhaoyang Yu · 3 Feb 2026
AOrchestra is a framework-agnostic agentic system that uses a tuple-based abstraction to dynamically create specialized task executors, achieving improved performance on complex benchmarks through automated agent creation and resource management.
Jason D. Lee, Yuanhe Zhang, Fanghui Liu · 2 Feb 2026
We present the first comprehensive Lean 4 formalization of statistical learning theory (SLT) grounded in empirical process theory. Our en-to-end formal infrastructure implement the missing contents in latest Lean library, including a complete development of Gaussian Lipschitz…
1 Feb 2026
Evaluating the quality of automatically generated text often relies on LLM-as-a-judge (LLM-judge) methods. While effective, these approaches are computationally expensive and require post-processing.
1 Feb 2026
We present a large-scale human evaluation benchmark for assessing cultural localisation in machine translation produced by state-of-the-art multilingual large language models (LLMs).