Papers

3,241 in the directory.

Browse AI research papers and methods. Search paper profiles, read recorded summaries and follow source links for the full research.

A–Z Recently synced

When Tokenizers Fail: Byte-Level Chunking for Zero-Shot Transfer to Low-Resource Languages taaft Subword tokenization hinders low-resource language processing by imposing frequency patterns from d… 2026-08-27 published
When Vision Meets Graphs: A Survey on Graph Reasoning and Learning taaft Graphs are a fundamental data structure underlying many problems in the natural and social sciences… 2026-09-03 published
Where can AI be used? Insights from a deep ontology of work activities Extropic Artificial intelligence (AI) is poised to profoundly reshape how work is executed and organized, bu… 2026-03-21 published
Where Do Deep-Research Agents Go Wrong? Span-Level Error Localization in Agent Trajectories taaft Deep-research agents solve tasks through long trajectories of search, tool use, evidence inspection… 2026-06-01 published
Where Do Multilingual Vision-Language Encoders Fail on Low-Resource Languages? taaft Recent multilingual vision--language encoders cover hundreds of languages in a single model, yet on… 2026-08-31 published
Where to Look Matters: Learning Influential Views for VLM-based 3D Visual Grounding taaft Recent zero-shot 3D visual grounding methods leverage vision-language models (VLMs) to localize obj… 2026-09-04 published
Where to Look Matters: On-Policy Self-Distillation for Long-Video Understanding taaft Vision-language models (VLMs) have made substantial progress in long-video understanding, with stan… 2026-08-26 published
Where vs What: Decomposing Structural and Content Failures in LLM-Generated Structured Outputs taaft Structured outputs such as JSON and tables are central to modern LLM-based systems, yet generation… 2026-08-26 published
Which LLM for Which Work? Budgeted Model Allocation under Uncertain Evaluation taaft A company with a fixed artificial intelligence (AI) budget must decide which large language model (… 2026-08-30 published
Which one is banana man? Evaluating vision-language models in multi-turn pragmatic interpretation taaft Flexible adaptation to context and shared pragmatic intuitions contribute to smooth human conversat… 2026-08-30 published
Who Drives the Probability Game of VLMs? A Temporal Causal Drive Evaluation Framework taaft Vision-language models (VLMs) are increasingly evaluated on complex image and video understanding t… 2026-09-02 published
Who Judges the Judges? A Chinese Safety QA Benchmark for Evaluating LLM Responses and Safety Judges taaft Safety benchmarks for large language models often assess the risk of a user query, although the out… 2026-09-01 published
Who Should Grade My Work? Student Perspectives on Transparent AI-Assisted Writing Assessment in Higher Education taaft The integration of GenAI tools into higher education assessment raises important questions about ho… 2026-09-04 published
Who Speaks for the Pruned? Visual Token Pruning as Coverage Optimization taaft Visual token pruning reduces the inference cost of vision-language models (VLMs), but most methods… 2026-09-02 published
Whole-Slide Image Analysis under Realistic Few-Shot Annotation Protocols taaft Automating the analysis of whole-slide images has high clinical value, since characterizing cancers… 2026-08-31 published
Whose record is this? Diagnosing and authorizing record use in personalized multimodal models taaft Contextualized visual personalization can retrieve a true record yet apply it to the wrong visual s… 2026-09-04 published
Why are all LLMs Obsessed with Japanese Culture? On the Hidden Cultural and Regional Biases of LLMs taaft LLMs have limitations when it comes to cultural coverage and competence, and in some cases, show sp… 2026-04-23 published
Why Didn't It Check? Unsupported Final Claims and Their Repair in Two Tool-Equipped Language Models taaft A language model with access to tools can commit to a final claim unsupported by the evidence it ha… 2026-08-27 published
Why Does Graph Learning Fail to Fully Benefit from a Text Teacher? taaft Graph neural networks (GNNs) are widely used to represent complex interactions and relationships am… 2026-08-26 published
Why Gated DeltaNet Survives 4-Bit Quantization: NVFP4 W4A4 for the Recurrent Half of a Hybrid 27B LLM taaft Hybrid LLMs pair softmax attention with linear-attention layers such as Gated DeltaNet (GDN), whose… 2026-09-03 published
Why ML-based cough models do not generalize: a systematic cross-dataset evaluation for tuberculosis screening taaft Cough acoustics are promising for non-invasive tuberculosis (TB) screening, yet whether machine lea… 2026-08-26 published
Why Multi-Layer Message Passing Works: Completeness Theory for Graph Neural Network Interatomic Potentials taaft We prove that the Hypergraph Neural Network, an invariant architecture with 3-body message passing,… 2026-09-01 published
Why we built an AI supercomputer in the cloud IBM Technical disclosure of IBM’s internal AI training cluster, detailing how they achieved bare-metal… IBM
Why We Care About Understanding: Competence through Predictive Compression taaft What is the relation between understanding and compression, and why does human understanding take s… 2026-09-04 published
WIDE: Wildcard Inference with Dynamic Expansion for Cross-Modal Generative Retrieval taaft Generative retrieval has demonstrated significant success by unifying representation learning and s… 2026-09-03 published
WiFlow: Estimating Optical Flow using WiFi Channel State Information taaft Knowing where and how fast objects are moving within a scene is important across various domains. U… 2026-09-02 published
WikiSkill: Compiling Agent Experience into Persistent Knowledge for Skill Evolution taaft Agent skills package specialized knowledge and workflows into reusable resources that extend AI age… 2026-08-27 published
WildSEEK: Evaluating Language Models for Information-Seeking taaft Language models are increasingly mediating information access to end users, urging a systematic eva… 2026-08-31 published
WilLaGS: Latent-Conditional 3D Appearance Fields for Robust Gaussian Splatting In-the-Wild taaft 3D Gaussian Splatting (3DGS) delivers real-time and high-fidelity rendering but remains challenged… 2026-08-28 published
WinoQueer-NL: Assessing Bias in Dutch Language Models toward LGBTQ+ Identities taaft While English language models have been widely examined for anti-queer bias, Dutch models remain un… 2026-09-02 published
Wireless Foundation Models: State-of-the-Art and Open Challenges taaft Wireless foundation models (WFMs) have emerged as a promising approach for learning reusable repres… 2026-09-04 published
WireSeg-32K: A Physics-Grounded Synthetic Dataset for Wire Instance Segmentation taaft Deformable linear objects such as wires and cables are difficult to segment because they are thin,… 2026-09-02 published
WiseSpec: Requirements-Driven Agents for Code Generation taaft Code generation aims to automatically generate source code from task requirements and has attracted… 2026-09-01 published
Witnesses Explain Anomalies taaft Unsupervised anomaly detection scores each point of an unlabelled, contaminated sample in a single… 2026-09-03 published
Word Importance Explains How Prompts Affect Language Model Outputs DataRobot The emergence of large language models (LLMs) has revolutionized numerous applications across indus… 2024-03-05 published
World Action Planner: Generalizable Decision-Making with Action-Conditioned World Models taaft Building generalizable agents for diverse applications remains a fundamental challenge. While imita… 2026-07-30 published