IP Library › Granted Patent US 12,651,125
Granted Patent B2
US 12,651,125 · App. 19/352,401 · Granted Jun 9, 2026

System and method for latent contextual threading in personalized dialogue using geometric manifold traversal

Inventor: Brian Galvin (Silverdale, WA)
Assignee: ATOMBEAM TECHNOLOGIES INC.
G06F40/35
View Patent ↗
Loading inventors, assignments & file history…
Monitor This Case
Get email alerts when status or documents change.
Order Certified Copies
Most orders are placed with the USPTO same day — all within 24 business hours.
Order via The Patent Place →
Pre-filled with this patent's details
Quick Facts
Patent No.
US 12,651,125
App. No.
19/352,401
Filed
Oct 7, 2025
Granted
Jun 9, 2026
Kind
B2
Art Unit
2658
USPC
704/9
Abstract

A system and methods for latent contextual threading for personalized dialogue through geometric manifold-based conversation management. The system maintains a personalized cognitive manifold as a geometric manifold in latent space that encodes user-specific dialogue patterns as navigable geometric structures. Multiple dialogue contexts are maintained as geometric trajectories within the manifold, with dialogue responses generated through manifold traversal rather than discrete context retrieval. A bidirectional adaptation system modifies the manifold's geometric structure based on user interactions. The system preserves dialogue continuity across session boundaries by serializing manifold geometry during session termination and restoring geometric positioning during session resumption. Dialogue coherence is evaluated through geometric analysis including curvature calculations and geodesic deviation measurements. The system maintains conversations through real-time manifold geometry modifications, providing dialogue experiences across session boundaries while maintaining contextual threading and personalized interaction patterns through geometric principles.

Claims (20)

1 . A computer system for latent contextual threading for personalized dialogue comprising a hardware memory, wherein the computer system is configured to execute software instructions stored on nontransitory machine-readable storage media that:

implement a personalized cognitive manifold as a Riemannian manifold with time-evolving metric tensors that encode dialogue preference strength through local curvature properties in latent space that encodes user-specific dialogue patterns as navigable geometric structures;

maintain multiple dialogue contexts as geometric trajectories within the personalized cognitive manifold, wherein dialogue responses are generated through manifold traversal comprising calculating geodesic paths through the personalized cognitive manifold and navigating along computed trajectories to maintain semantic coherence, rather than discrete context retrieval; and

modify the geometric structure of the personalized cognitive manifold based on user interactions by processing user feedback through geometric deformation algorithms that implement curvature flow equations to evolve metric tensor properties of the personalized cognitive manifold based on user interaction patterns, to improve dialogue personalization through bidirectional adaptation;

wherein dialogue continuity is maintained through geometric trajectory following within the personalized cognitive manifold, enabling persistent conversations that adapt to user preferences through real-time manifold geometry modifications.

2 . The computer system of claim 1 , wherein maintaining multiple dialogue contexts comprises implementing a contextual thread manager that assigns unique thread identifiers and tracks manifold coordinate positions for each dialogue trajectory.

3 . The computer system of claim 1 , wherein modifying the geometric structure comprises computing curvature adjustments based on user interaction patterns and implementing geometric flow processes that reshape manifold properties while preserving mathematical consistency.

4 . The computer system of claim 1 , wherein the computer system is further configured to execute software instructions that preserve dialogue trajectory states across session boundaries by serializing manifold geometry and restoring geometric positioning during session resumption.

5 . The computer system of claim 1 , wherein the computer system is further configured to execute software instructions that evaluate dialogue coherence through geometric analysis including Ricci curvature calculations and geodesic deviation measurements within the manifold.

6 . The computer system of claim 1 , wherein the bidirectional adaptation comprises analyzing user engagement patterns and implementing real-time metric tensor modifications that adjust manifold distance relationships to reflect learned user preferences.

7 . A method for latent contextual threading for personalized dialogue comprising the steps of:

implementing a personalized cognitive manifold as a Riemannian manifold with time-evolving metric tensors that encode dialogue preference strength through local curvature properties in latent space that encodes user-specific dialogue patterns as navigable geometric structures;

maintaining multiple dialogue contexts as geometric trajectories within the personalized cognitive manifold, wherein dialogue responses are generated through manifold traversal comprising calculating geodesic paths through the personalized cognitive manifold and navigating along computed trajectories to maintain semantic coherence, rather than discrete context retrieval; and

modifying the geometric structure of the personalized cognitive manifold based on user interactions by processing user feedback through geometric deformation algorithms that implement curvature flow equations to evolve metric tensor properties of the personalized cognitive manifold based on user interaction patterns, to improve dialogue personalization through bidirectional adaptation;

wherein dialogue continuity is maintained through geometric trajectory following within the personalized cognitive manifold, enabling persistent conversations that adapt to user preferences through real-time manifold geometry modifications.

8 . The method of claim 7 , wherein maintaining multiple dialogue contexts comprises implementing a contextual thread manager that assigns unique thread identifiers and tracks manifold coordinate positions for each dialogue trajectory.

9 . The method of claim 7 , wherein modifying the geometric structure comprises computing curvature adjustments based on user interaction patterns and implementing geometric flow processes that reshape manifold properties while preserving mathematical consistency.

10 . The method of claim 7 , further comprising preserving dialogue trajectory states across session boundaries by serializing manifold geometry and restoring geometric positioning during session resumption.

11 . The method of claim 7 , further comprising evaluating dialogue coherence through geometric analysis including Ricci curvature calculations and geodesic deviation measurements within the manifold.

12 . The method of claim 7 , wherein the bidirectional adaptation comprises analyzing user engagement patterns and implementing real-time metric tensor modifications that adjust manifold distance relationships to reflect learned user preferences.

Assignments (1)
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Dec 2, 2025
From: GALVIN, BRIAN
To: ATOMBEAM TECHNOLOGIES INC.
Reel/Frame 073093/0016 →
Continuity (42)
Continuation In Part 19339302 · Sep 25, 2025
Continuation In Part 19294125 · Aug 7, 2025
Continuation In Part 19203069 · Jun 3, 2025
Continuation In Part 19205960 · May 12, 2025
Continuation In Part 19060794 · Feb 24, 2025
Continuation In Part 19044546 · Feb 3, 2025
Continuation In Part 19026276 · Jan 16, 2025
Continuation In Part 18928022 · Oct 26, 2024
Continuation In Part 18919417 · Oct 17, 2024
Continuation In Part 18918077 · Oct 17, 2024
Continuation In Part 18737906 · Jun 7, 2024
Continuation In Part 18736498 · Jun 6, 2024
Continuation In Part 19178873 · Apr 15, 2025
Continuation In Part 19177611 · Apr 13, 2025
Continuation In Part 19051193 · Feb 12, 2025
Continuation In Part 19329533 · Sep 15, 2025
Continuation In Part 19329369 · Sep 15, 2025
Continuation In Part 19328199 · Sep 14, 2025
Continuation In Part 19328179 · Sep 14, 2025
Continuation In Part 19328103 · Sep 13, 2025
Continuation In Part 19326730 · Sep 12, 2025
Continuation In Part 19321173 · Sep 6, 2025
Continuation In Part 19284115 · Jul 29, 2025
Continuation In Part 19051193 · Feb 12, 2025
Continuation In Part 19245366 · Jun 22, 2025
Continuation In Part 19204525 · May 10, 2025
Continuation In Part 19192215 · Apr 28, 2025
Continuation In Part 18972797 · Dec 6, 2024
Continuation In Part 18648340 · Apr 27, 2024
Continuation In Part 18427716 · Jan 30, 2024
Continuation In Part 18410980 · Jan 11, 2024
Continuation In Part 18537728 · Dec 12, 2023
Provisional Application 63651359 · May 23, 2024
Provisional Application 63847408 · Jul 20, 2025
Provisional Application 63847082 · Jul 19, 2025
Provisional Application 63847091 · Jul 19, 2025
Provisional Application 63847096 · Jul 19, 2025
Provisional Application 63847101 · Jul 19, 2025
Provisional Application 63847889 · Jul 21, 2025
Provisional Application 63848101 · Jul 21, 2025
Provisional Application 63848202 · Jul 21, 2025
Related Publication 20260037743A1 · Feb 5, 2026
References Cited (4)
US 20190147349A1 · Ng · 2019 [cited by examiner]
US 20230401262A1 · Orús Lacort · 2023 [cited by examiner]
US 20240135283A1 · Sahoo · 2024 [cited by examiner]
Gim, In, et al; “Prompt Cache: Modular Attention Reuse for Low-Latency Inference”, arXiv:2311.04934v2, Apr. 2024. [cited by applicant]