reasoning
The process of drawing conclusions or inferences based on knowledge, data, and logical rules. In AI, reasoning often involves using learned representations to make logical deductions from available information.
- AVerImaTeC: A Dataset for Automatic Verification of Image-Text Claims with Evidence from the Web
- AgentBreeder: Mitigating the AI Safety Risks of Multi-Agent Scaffolds via Self-Improvement
- Anchored Diffusion Language Model
- CHOICE: Benchmarking the Remote Sensing Capabilities of Large Vision-Language Models
- CoC-VLA: Delving into Adversarial Domain Transfer for Explainable Autonomous Driving via Chain-of-Causality Visual-Language-Action Model
- Contextual Integrity in LLMs via Reasoning and Reinforcement Learning
- DreamPRM: Domain-reweighted Process Reward Model for Multimodal Reasoning
- DreamVLA: A Vision-Language-Action Model Dreamed with Comprehensive World Knowledge
- FailureSensorIQ: A Multi-Choice QA Dataset for Understanding Sensor Relationships and Failure Modes
- Foundation Models for Scientific Discovery: From Paradigm Enhancement to Paradigm Transition
- GFM-RAG: Graph Foundation Model for Retrieval Augmented Generation
- Geometry of Decision Making in Language Models
- Hogwild! Inference: Parallel LLM Generation via Concurrent Attention
- InfoChartQA: A Benchmark for Multimodal Question Answering on Infographic Charts
- KVzip: Query-Agnostic KV Cache Compression with Context Reconstruction
- KVzip: Query-Agnostic KV Cache Compression with Context Reconstruction
- MMTU: A Massive Multi-Task Table Understanding and Reasoning Benchmark
- MR. Video: MapReduce as an Effective Principle for Long Video Understanding
- Matryoshka Pilot: Learning to Drive Black-Box LLMs with LLMs
- OmniBench: Towards The Future of Universal Omni-Language Models
- ROVER: Recursive Reasoning Over Videos with Vision-Language Models for Embodied Tasks
- ReMA: Learning to Meta-Think for LLMs with Multi-agent Reinforcement Learning
- Reasoning is Periodicity? Improving Large Language Models Through Effective Periodicity Modeling
- RvLLM: LLM Runtime Verification with Domain Knowledge
- Sloth: scaling laws for LLM skills to predict multi-benchmark performance across families
- SpatialReasoner: Towards Explicit and Generalizable 3D Spatial Reasoning
- The Lighthouse of Language: Enhancing LLM Agents via Critique-Guided Improvement
- Transformers Provably Learn Chain-of-Thought Reasoning with Length Generalization
- VLM-R³: Region Recognition, Reasoning, and Refinement for Enhanced Multimodal Chain-of-Thought
- WearVQA: A Visual Question Answering Benchmark for Wearables in Egocentric Authentic Real-world scenarios
- Wide-Horizon Thinking and Simulation-Based Evaluation for Real-World LLM Planning with Multifaceted Constraints
- World Models Should Prioritize the Unification of Physical and Social Dynamics