EchoGuard: An Agentic Framework with Knowledge-Graph Memory for Detecting Manipulative Communication in Longitudinal Dialogue
#EchoGuard #manipulative communication #knowledge-graph #longitudinal dialogue #agentic framework #detection #memory system
📌 Key Takeaways
- EchoGuard is a new framework designed to detect manipulative communication in ongoing dialogues.
- It uses an agentic approach combined with a knowledge-graph memory system.
- The framework is specifically tailored for analyzing longitudinal dialogue data.
- It aims to identify patterns of manipulation over extended conversational interactions.
📖 Full Retelling
arXiv:2603.04815v1 Announce Type: new
Abstract: Manipulative communication, such as gaslighting, guilt-tripping, and emotional coercion, is often difficult for individuals to recognize. Existing agentic AI systems lack the structured, longitudinal memory to track these subtle, context-dependent tactics, often failing due to limited context windows and catastrophic forgetting. We introduce EchoGuard, an agentic AI framework that addresses this gap by using a Knowledge Graph (KG) as the agent's c
🏷️ Themes
AI Detection, Communication Analysis
Entity Intersection Graph
No entity connections available yet for this article.
Original Source
--> Computer Science > Artificial Intelligence arXiv:2603.04815 [Submitted on 5 Mar 2026] Title: EchoGuard: An Agentic Framework with Knowledge-Graph Memory for Detecting Manipulative Communication in Longitudinal Dialogue Authors: Ratna Kandala , Niva Manchanda , Akshata Kishore Moharir , Ananth Kandala View a PDF of the paper titled EchoGuard: An Agentic Framework with Knowledge-Graph Memory for Detecting Manipulative Communication in Longitudinal Dialogue, by Ratna Kandala and 3 other authors View PDF Abstract: Manipulative communication, such as gaslighting, guilt-tripping, and emotional coercion, is often difficult for individuals to recognize. Existing agentic AI systems lack the structured, longitudinal memory to track these subtle, context-dependent tactics, often failing due to limited context windows and catastrophic forgetting. We introduce EchoGuard, an agentic AI framework that addresses this gap by using a Knowledge Graph as the agent's core episodic and semantic memory. EchoGuard employs a structured Log-Analyze-Reflect loop: (1) users log interactions, which the agent structures as nodes and edges in a personal, episodic KG (capturing events, emotions, and speakers); (2) the system executes complex graph queries to detect six psychologically-grounded manipulation patterns (stored as a semantic KG 3) an LLM generates targeted Socratic prompts grounded by the subgraph of detected patterns, guiding users toward self-discovery. This framework demonstrates how the interplay between agentic architectures and Knowledge Graphs can empower individuals in recognizing manipulative communication while maintaining personal autonomy and safety. We present the theoretical foundation, framework design, a comprehensive evaluation strategy, and a vision to validate this approach. Subjects: Artificial Intelligence (cs.AI) Cite as: arXiv:2603.04815 [cs.AI] (or arXiv:2603.04815v1 [cs.AI] for this version) https://doi.org/10.48550/arXiv.2603.04815 Focus to learn more arXi...
Read full article at source