Complex Reasoning

Complex reasoning in artificial intelligence focuses on developing models capable of multi-step, logical inference and problem-solving, mirroring human cognitive abilities. Current research emphasizes improving large language models (LLMs) through techniques like chain-of-thought prompting, retrieval-augmented generation (RAG), and the integration of symbolic reasoning with neural networks, often incorporating multi-modal data (e.g., visual and textual information). These advancements are significant for enhancing the reliability and applicability of AI systems across diverse fields, including autonomous driving, robotics, and scientific discovery, by enabling more robust and accurate decision-making in complex scenarios.

552papers

Papers

May 23, 2025

Bridging Supervised Learning and Reinforcement Learning in Math Reasoning
Self Feedback Reinforcement Learning Supervised Learning Complex Reasoning Binary Classifier Implicit Policy Level Mathematics
CXReasonBench: A Benchmark for Evaluating Structured Diagnostic Reasoning in Chest X-rays
Chest X Ray New Benchmark Large Vision Language Model Visual Grounding Complex Reasoning Visual Input Medical Information Mart for Intensive Diagnostic Model Diagnostic Reasoning
Reward Model Generalization for Compute-Aware Test-Time Reasoning
Test Time Adaptation Test Time Optimization Complex Reasoning Reward Function Inference Cost Reasoning Performance
Stepwise Reasoning Checkpoint Analysis: A Test Time Scaling Method to Enhance LLMs' Reasoning
Reasoning Path Intermediate Checkpoint Complex Reasoning Reasoning Step Step by Step Reasoning
Don't Overthink it. Preferring Shorter Thinking Chains for Improved LLM Reasoning
LLM Generated Reasoning ChAin Complex Reasoning Large Language Model Complex Reasoning Task LLM Inference
Rethinking the Sampling Criteria in Reinforcement Learning for LLM Reasoning: A Competence-Difficulty Alignment Perspective
Dynamic Sampling Model Capability Complex Reasoning Medical LLM Reinforcement Learning Large Language Model
HoloLLM: Multisensory Foundation Model for Language-Grounded Human Sensing and Reasoning
Language Grounding Vision Language Model Complex Reasoning Multimodal Large Language Model Human Sensing Multimodal Foundation Model
Controlled Agentic Planning & Reasoning for Mechanism Synthesis
Symbolic Regression Complex Reasoning Mechanism Synthesis Large Language Model Agent Planning
Reasoning Meets Personalization: Unleashing the Potential of Large Reasoning Model for Personalized Generation
Complex Reasoning Reasoning System Personalized Generation Reasoning Capability Full Potential Concept Personalization Hierarchical Reasoning Language Model
On the Design of KL-Regularized Policy Gradient Algorithms for LLM Reasoning
Product Design Medical LLM Policy Regularization Online Reinforcement Learning Regularized Reinforcement Learning Complex Reasoning Policy Gradient
CReSt: A Comprehensive Benchmark for Retrieval-Augmented Generation with Complex Reasoning over Structured Documents
Complex Reasoning RAG Based Comprehensive Benchmark Structured Document Retrieval Augmented Generation
From Reasoning to Generalization: Knowledge-Augmented LLMs for ARC Benchmark
Reasoning Corpus New Benchmark LLM Reasoning Complex Reasoning Medical LLM Strong Generalization Abstract Reasoning
MARCO: Meta-Reflection with Cross-Referencing for Code Reasoning
Task Complexity Logical Reasoning Complex Reasoning Reasoning Path
Self-Training Large Language Models with Confident Reasoning
Reasoning Path Complex Reasoning Language Model Self Supervised Large Language Model
Language Matters: How Do Multilingual Input and Reasoning Paths Affect Large Reasoning Models?
Language Input Complex Reasoning Reasoning Task Natural Language Reasoning Large Knowledge Model Complex Reasoning Task Reasoning Path
Misaligning Reasoning with Answers -- A Framework for Assessing LLM CoT Robustness
Complex Reasoning Native Robustness Novel Evaluation LLM Reasoning New Framework Top Two Answer

Complex Reasoning

Papers

Bridging Supervised Learning and Reinforcement Learning in Math Reasoning

CXReasonBench: A Benchmark for Evaluating Structured Diagnostic Reasoning in Chest X-rays

Reward Model Generalization for Compute-Aware Test-Time Reasoning

Stepwise Reasoning Checkpoint Analysis: A Test Time Scaling Method to Enhance LLMs' Reasoning

Don't Overthink it. Preferring Shorter Thinking Chains for Improved LLM Reasoning

Rethinking the Sampling Criteria in Reinforcement Learning for LLM Reasoning: A Competence-Difficulty Alignment Perspective

HoloLLM: Multisensory Foundation Model for Language-Grounded Human Sensing and Reasoning

Controlled Agentic Planning & Reasoning for Mechanism Synthesis

Reasoning Meets Personalization: Unleashing the Potential of Large Reasoning Model for Personalized Generation

On the Design of KL-Regularized Policy Gradient Algorithms for LLM Reasoning

CReSt: A Comprehensive Benchmark for Retrieval-Augmented Generation with Complex Reasoning over Structured Documents

From Reasoning to Generalization: Knowledge-Augmented LLMs for ARC Benchmark

MARCO: Meta-Reflection with Cross-Referencing for Code Reasoning

Self-Training Large Language Models with Confident Reasoning

Language Matters: How Do Multilingual Input and Reasoning Paths Affect Large Reasoning Models?

Misaligning Reasoning with Answers -- A Framework for Assessing LLM CoT Robustness

Longer Context, Deeper Thinking: Uncovering the Role of Long-Context Ability in Reasoning

AdaReasoner: Adaptive Reasoning Enables More Flexible Thinking

Select2Reason: Efficient Instruction-Tuning Data Selection for Long-CoT Reasoning

Effective Reinforcement Learning for Reasoning in Language Models