Goal-Oriented Reasoning for RAG-based Memory in Conversational Agentic LLM Systems

2026-05-12

AI Summary

This work addresses the incoherent behavior of large language model agents in long-horizon dialogues, which often stems from inadequate memory mechanisms. While existing retrieval-augmented generation (RAG) approaches rely on semantic similarity between user utterances and stored memories, they struggle with complex tasks requiring multi-hop or implicit commonsense reasoning. To overcome this limitation, the paper introduces Goal-Mem, a novel framework that incorporates goal-directed backward chaining into RAG for the first time. It parses user utterances into high-level goals, recursively decomposes them into atomic subgoals, and performs targeted memory retrieval accordingly; when subgoals cannot be fulfilled, the system iteratively identifies missing information. By integrating natural language with logically formalized reasoning, Goal-Mem enables verifiable and highly expressive inference. Experiments demonstrate that Goal-Mem significantly outperforms nine strong baselines on two benchmarks, achieving particularly notable gains in multi-hop and implicit commonsense reasoning tasks.

Abstract

LLM-based conversational AI agents struggle to maintain coherent behavior over long horizons due to limited context. While RAG-based approaches are increasingly adopted to overcome this limitation by storing interactions in external memory modules and performing retrieval from them, their effectiveness in answering challenging questions (e.g., multi-hop, commonsense) ultimately depends on the agent's ability to reason over the retrieved information. However, existing methods typically retrieve memory based on semantic similarity to the raw user utterance, which lacks explicit reasoning about missing intermediate facts and often returns evidence that is irrelevant or insufficient for grounded reasoning. In this work, we introduce Goal-Mem, a goal-oriented reasoning framework for RAG-based agentic memory that performs explicit backward chaining from the user's utterance as a goal. Rather than progressively expanding from retrieved context, Goal-Mem decomposes each goal into atomic subgoals, performs targeted memory retrieval to satisfy each subgoal, and iteratively identifies what information from memory should be retrieved when intermediate goals cannot be resolved. We formalize this process in Natural Language Logic, a logical system that combines the verifiability of reasoning provided by FOL with the expressivity of natural language. Through extensive experiments on two datasets and comparing to nine strong memory baselines, we show that Goal-Mem consistently improves performance, particularly on tasks requiring multi-hop reasoning and implicit inference.

Problem

• Conversational Agentic LLM
• RAG-based Memory
• Goal-Oriented Reasoning
• Multi-hop Reasoning
• Memory Retrieval

Innovation

• goal-oriented reasoning
• RAG-based memory
• backward chaining
• Natural Language Logic
• multi-hop reasoning