multi-step reasoning
The ability of an AI system to perform reasoning tasks that require several sequential steps or logical deductions to reach conclusions.
- A Little Depth Goes a Long Way: The Expressive Power of Log-Depth Transformers
- Atomic Thinking of LLMs: Decoupling and Exploring Mathematical Reasoning Abilities
- Auditing Meta-Cognitive Hallucinations in Reasoning Large Language Models
- Automated Model Discovery via Multi-modal & Multi-step Pipeline
- Beyond Oracle: Verifier-Supervision for Instruction Hierarchy in Reasoning and Instruction-Tuned LLMs
- BioReason: Incentivizing Multimodal Biological Reasoning within a DNA-LLM Model
- Causal-R: A Causal-Reasoning Geometry Problem Solver for Optimized Solution Exploration
- CoRe: Benchmarking LLMs’ Code Reasoning Capabilities through Static Analysis Tasks
- CoT Information: Improved Sample Complexity under Chain-of-Thought Supervision
- Conformal Prediction Beyond the Seen: A Missing Mass Perspective for Uncertainty Quantification in Generative Models
- From Style to Facts: Mapping the Boundaries of Knowledge Injection with Finetuning
- GPO: Learning from Critical Steps to Improve LLM Reasoning
- Language Models can Self-Improve at State-Value Estimation for Better Search
- Learning to Reason under Off-Policy Guidance
- LogicTree: Improving Complex Reasoning of LLMs via Instantiated Multi-step Synthetic Logical Data
- MMAR: A Challenging Benchmark for Deep Reasoning in Speech, Audio, Music, and Their Mix
- Multi-head Transformers Provably Learn Symbolic Multi-step Reasoning via Gradient Descent
- Once Upon an Input: Reasoning via Per-Instance Program Synthesis
- PHYBench: Holistic Evaluation of Physical Perception and Reasoning in Large Language Models
- PlanU: Large Language Model Reasoning through Planning under Uncertainty
- ReCAP: Recursive Context-Aware Reasoning and Planning for Large Language Model Agents
- Reasoning Beyond Points: A Visual Introspective Approach for Few-Shot 3D Segmentation
- SEEA-R1: Tree-Structured Reinforcement Fine-Tuning for Self-Evolving Embodied Agents
- SpatialReasoner: Towards Explicit and Generalizable 3D Spatial Reasoning
- Think or Not? Exploring Thinking Efficiency in Large Reasoning Models via an Information-Theoretic Lens
- WebDancer: Towards Autonomous Information Seeking Agency