AI & ML
IntentWeaver: Sustaining Coherence across Evolving Intents in Human-AI Co-Writing
arXiv cs.HC0 views
arXiv:2611.08214v1 Announce Type: new Abstract: Writers hold intents at many levels, from overarching goals to concrete wording, and evolve them by periodically popping back up from the text to higher-level intents. Generative AI co-writing disrupts this practice. Dense drafts capture attention and trap writers in local editing; higher-level intents stall while lower-level ones drift from them unnoticed, a breakdown we term intent incoherence. Our formative study traced this breakdown from attention capture through silent drift to failed reconciliation. We present IntentWeaver, which supports intent evolution while sustaining coherence in human-AI co-writing. It derives an intent hierarchy from the draft and writer interactions, evaluates each intent's fulfillment and cross-level coherence at every change, and propagates consequences across intent levels. In a within-subjects study (N=24) against a baseline, participants using IntentWeaver revisited and evolved higher-level intents more often, with fewer unnoticed deviations, and produced texts better realizing them. We further discuss design implications for emergent intents.
Source:https://arxiv.org/abs/2611.08214PreviousReflective Nature Experiences in Virtual RealityNextControlAnything: Generative UI and On-Demand Device Integration for Mixed-Reality Hand Interaction by a Coding Agent
Related posts
How to Convert Any Webpage to Clean Markdown for RAG & Save 90% LLM Tokens in Node.js
0dev.to
Calibration‑free top‑k MoE selector saves tuning
0dev.to
How to Evaluate RAG Retrieval in 2026: Question Sets, Metrics That Mislead, and a Practical Recipe
0dev.to
Reading a Competitor's Pricing Page Like an Engineer
0dev.to
Day 4: Adding VLAN support
0dev.to

Browser-local image tools: how to check the upload boundary
0dev.to