The field of large language models (LLMs) is rapidly advancing, with a focus on improving complex decision-making capabilities. Recent research has explored the use of LLMs in various applications, including marketing, inventory management, and legal reasoning. A key trend is the development of multi-agent systems that enable LLMs to collaborate and make decisions in a more robust and reliable manner. Another area of focus is the integration of legal logic into deep learning models, which has shown promise in improving the accuracy and interpretability of legal decision-making. Noteworthy papers in this area include "Towards Reliable Multi-Agent Systems for Marketing Applications via Reflection, Memory, and Planning" and "LegalΔ: Enhancing Legal Reasoning in LLMs via Reinforcement Learning with Chain-of-Thought Guided Information Gain". These papers demonstrate the potential of LLMs to improve decision-making in complex domains and highlight the need for further research in this area.
Advances in Large Language Models for Complex Decision-Making
Sources
Towards Reliable Multi-Agent Systems for Marketing Applications via Reflection, Memory, and Planning
AIM-Bench: Evaluating Decision-making Biases of Agentic LLM as Inventory Manager
AgentCDM: Enhancing Multi-Agent Collaborative Decision-Making via ACH-Inspired Structured Reasoning
Legal$\Delta$: Enhancing Legal Reasoning in LLMs via Reinforcement Learning with Chain-of-Thought Guided Information Gain
Incorporating Legal Logic into Deep Learning: An Intelligent Approach to Probation Prediction
Wisdom of the Crowd: Reinforcement Learning from Coevolutionary Collective Feedback
ReaLM: Reflection-Enhanced Autonomous Reasoning with Small Language Models
SSPO: Self-traced Step-wise Preference Optimization for Process Supervision and Reasoning Compression
ToolACE-MT: Non-Autoregressive Generation for Agentic Multi-Turn Interaction
Atom-Searcher: Enhancing Agentic Deep Research via Fine-Grained Atomic Thought Reward
Can Large Models Teach Student Models to Solve Mathematical Problems Like Human Beings? A Reasoning Distillation Method via Multi-LoRA Interaction
AutoBnB-RAG: Enhancing Multi-Agent Incident Response with Retrieval-Augmented Generation
Chain-of-Agents: End-to-End Agent Foundation Models via Multi-Agent Distillation and Agentic RL
A Risk Manager for Intrusion Tolerant Systems: Enhancing HAL 9000 with New Scoring and Data Sources
Datarus-R1: An Adaptive Multi-Step Reasoning LLM for Automated Data Analysis
Conflicting Scores, Confusing Signals: An Empirical Study of Vulnerability Scoring Systems
Sycophancy under Pressure: Evaluating and Mitigating Sycophantic Bias via Adversarial Dialogues in Scientific QA
The Collaboration Paradox: Why Generative AI Requires Both Strategic Intelligence and Operational Stability in Supply Chain Management
MultiFuzz: A Dense Retrieval-based Multi-Agent System for Network Protocol Fuzzing
Who Sees What? Structured Thought-Action Sequences for Epistemic Reasoning in LLMs
HERAKLES: Hierarchical Skill Compilation for Open-ended LLM Agents
Disentangling the Drivers of LLM Social Conformity: An Uncertainty-Moderated Dual-Process Mechanism
When Machine Learning Meets Vulnerability Discovery: Challenges and Lessons Learned
LLMs and Agentic AI in Insurance Decision-Making: Opportunities and Challenges For Africa
A Practical Guideline and Taxonomy to LLVM's Control Flow Integrity
From Bits to Boardrooms: A Cutting-Edge Multi-Agent LLM Framework for Business Excellence
Transduction is All You Need for Structured Data Workflows