List рдкрд░ рд╡рд╛рдкрд╕ рдЬрд╛рдПрдБ
ЁЯУДPapers

Inflak: A Unified Orchestration Architecture for Human-Agent Interaction

Gao L. et al.
2026-10-15

рдкрд░рд┐рдЪрдп

October 2026 paper separating runtime human-agent interaction into routing, planning, design, and rendering. Inflak implements these layers as composable agent skills, letting task state determine when to request input and which generated interface to present.

рд╕рд╛рд░рд╛рдВрд╢

Why Inflak Adds New Signal:

1. Explicit layer contracts: Separates task progression, interaction choices, widget specifications, and rendering through structured handoffs
2. Concrete implementation: Uses GitHub Copilot plugins and a web frontend, with 13 interaction paradigms and combined chat/canvas workflows
3. Extension evidence: Compares adding interactions through layered skills with a loosely structured implementation
4. Evaluation boundary: Measures layer compliance and latency on two host models; the technical evaluation does not assess final widget quality or demonstrate broad user benefit

Tags

generative-uiinteraction-orchestrationagent-skillsruntime-uimixed-initiative
рдореВрд▓ рд▓реЗрдЦ рдкрдврд╝реЗрдВ

рдорд┐рд▓рддреЗ-рдЬреБрд▓рддреЗ рд▓реЗрдЦ

ЁЯУДPapers

Atelier: From Semantic Editing Intentions to Persistent Parametric Controls for AI-Assisted Image Editing

October 2026 HCI paper on adaptive controls for image editing. Atelier combines image context and a user's editing intent to compose persistent, adjustable widgets, exposing proposed parameters for refinement instead of returning only a finished image.

generative-uiimage-editingadaptive-controls
рд▓реЗрдЦрдХ Amin R. M. et al. ┬╖ 2026-10-14
рдФрд░ рдкрдврд╝реЗрдВ
ЁЯУДPapers

When Interfaces Speak: Data-Aware Generative UI Harness for Active Interaction

October 2026 paper pairing database tool execution with task-specific frontend generation. GenUI-Harness adds sandboxed interaction rollouts and reward auditing, while UI-TAU Bench tests whether generated interfaces help agents collect input and complete database-backed tasks across ten domains.

generative-uidata-grounded-uiui-tau-bench
рд▓реЗрдЦрдХ Li X. et al. ┬╖ 2026-10-08
рдФрд░ рдкрдврд╝реЗрдВ
ЁЯУДPapers

Design Creativity Bench: Measuring creativity in LLM-Generated UI

October 2026 Kombai paper evaluating originality, variation across product domains, and requirement satisfaction in LLM-generated interfaces. It compares 2,400 HTML designs from 15 models across 160 prompts with human reference designs, separating acceptable output from diverse design.

ui-generationdesign-diversitycreativity-evaluation
рд▓реЗрдЦрдХ Rusia A. et al. ┬╖ 2026-10-08
рдФрд░ рдкрдврд╝реЗрдВ

Generative UI рдФрд░ Intelligent UI resources рдХрд╛ curated рд╕рдВрдЧреНрд░рд╣ред

рдмрдирд╛рдпрд╛ рдЧрдпрд╛ тЭдя╕П community рджреНрд╡рд╛рд░рд╛