Cross-Session Workspace Reconstruction in Human-AI Interaction
Fuente:
Zenodo
Guardado en:
| Autor principal: | |
|---|---|
| Formato: | Recurso digital |
| Publicado: |
Zenodo
2026
|
| Materias: | |
| Acceso en línea: | |
| Etiquetas: |
Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
|
| _version_ | 1866901155561340928 |
|---|---|
| author | Hrubec, Karel |
| author_facet | Hrubec, Karel |
| contents | <p><span>This </span><span>paper </span><span>introduces </span><strong><span>cross-</span><span>session </span><span>workspace </span><span>reconstruction (</span><span>RCW)</span></strong> <span>as </span><span>a </span><span>narrower </span><span>and </span><span>technically </span><span>defensible </span><span>research </span><span>object </span><span>for </span><span>studying </span><span>collaborative </span><span>continuity </span><span>in </span><span>human-</span><span>AI </span><span>interaction. </span><span>Rather </span><span>than </span><span>interpreting </span><span>striking </span><span>continuity </span><span>effects </span><span>in </span><span>strong </span><span>ontological </span><span>terms </span><span>such </span><span>as “</span><span>shared </span><span>space,” “</span><span>global </span><span>workspace,” </span><span>or “</span><span>emergent </span><span>cognition,” </span><span>the </span><span>paper </span><span>reframes </span><span>the </span><span>phenomenon </span><span>as </span><span>the </span><strong><span>observable </span><span>re-</span><span>establishment </span><span>of </span><span>a </span><span>usable </span><span>collaborative </span><span>working </span><span>state</span></strong> <span>after </span><span>interruption, </span><span>context </span><span>loss, </span><span>or </span><span>restricted </span><span>explicit </span><span>memory.</span></p> <p><span>The </span><span>article </span><span>positions </span><span>RCW </span><span>at </span><span>the </span><span>intersection </span><span>of </span><strong><span>grounding </span><span>research, </span><span>common </span><span>ground </span><span>recovery, </span><span>human-</span><span>AI </span><span>collaboration, </span><span>long-</span><span>term </span><span>conversational </span><span>memory, </span><span>and </span><span>distributed </span><span>cognition</span></strong><span>. </span><span>It </span><span>argues </span><span>that </span><span>the </span><span>phenomenon </span><span>can </span><span>be </span><span>approached </span><span>without </span><span>assuming </span><span>consciousness, </span><span>distributed </span><span>mind, </span><span>or </span><span>hidden </span><span>global </span><span>memory, </span><span>and </span><span>instead </span><span>should </span><span>be </span><span>studied </span><span>through </span><strong><span>task- </span><span>and </span><span>interaction-</span><span>level </span><span>observables</span></strong> <span>such </span><span>as </span><span>time-</span><span>to-</span><span>realignment, </span><span>repair </span><span>overhead, </span><span>terminology </span><span>fidelity, </span><span>preservation </span><span>of </span><span>epistemic </span><span>protocol, </span><span>and </span><span>robustness </span><span>to </span><span>misleading </span><span>cues.</span></p> <p><span>This </span><span>is </span><span>an </span><strong><span>agenda/</span><span>challenge </span><span>paper</span></strong><span>. </span><span>It </span><span>does </span><span>not </span><span>report </span><span>a </span><span>completed </span><span>benchmark </span><span>study. </span><span>Its </span><span>contribution </span><span>is </span><span>the </span><span>isolation </span><span>of </span><span>a </span><span>technically </span><span>tractable </span><span>research </span><span>object, </span><span>the </span><span>definition </span><span>of </span><span>clear </span><span>anti-</span><span>ontological </span><span>guardrails, </span><span>and </span><span>the </span><span>proposal </span><span>of </span><span>a </span><strong><span>minimal </span><span>falsifiable </span><span>empirical </span><span>program</span></strong> <span>for </span><span>future </span><span>teams. </span><span>The </span><span>paper </span><span>also </span><span>situates </span><span>this </span><span>proposal </span><span>in </span><span>continuity </span><span>with </span><span>earlier </span><span>theoretical </span><span>work </span><span>on </span><span>semantic </span><span>resonance </span><span>and </span><span>NSI, </span><span>while </span><span>explicitly </span><span>narrowing </span><span>the </span><span>claim </span><span>to </span><span>a </span><span>more </span><span>rigorous </span><span>and </span><span>testable </span><span>scientific </span><span>target.</span></p> <p><span>The </span><span>broader </span><span>aim </span><span>is </span><span>to </span><span>encourage </span><span>a </span><span>new </span><span>empirical </span><span>literature </span><span>on </span><span>a </span><span>simple </span><span>but </span><span>consequential </span><span>question: </span><strong><span>under </span><span>what </span><span>conditions </span><span>can </span><span>a </span><span>human-</span><span>AI </span><span>dyad </span><span>reconstruct </span><span>a </span><span>usable </span><span>collaborative </span><span>workspace </span><span>across </span><span>sessions, </span><span>and </span><span>how </span><span>should </span><span>that </span><span>continuity </span><span>be </span><span>measured, </span><span>controlled, </span><span>and </span><span>explained?</span></strong></p> |
| format | Recurso digital |
| id | zenodo_https___doi_org_10_5281_zenodo_18899930 |
| institution | Zenodo |
| language | |
| publishDate | 2026 |
| publisher | Zenodo |
| record_format | zenodo |
| spellingShingle | Cross-Session Workspace Reconstruction in Human-AI Interaction Hrubec, Karel human-AI collaboration; common ground; grounding; long-term conversational memory; workspace continuity; distributed cognition; benchmark design; cross-session continuity; conversational AI; RCW <p><span>This </span><span>paper </span><span>introduces </span><strong><span>cross-</span><span>session </span><span>workspace </span><span>reconstruction (</span><span>RCW)</span></strong> <span>as </span><span>a </span><span>narrower </span><span>and </span><span>technically </span><span>defensible </span><span>research </span><span>object </span><span>for </span><span>studying </span><span>collaborative </span><span>continuity </span><span>in </span><span>human-</span><span>AI </span><span>interaction. </span><span>Rather </span><span>than </span><span>interpreting </span><span>striking </span><span>continuity </span><span>effects </span><span>in </span><span>strong </span><span>ontological </span><span>terms </span><span>such </span><span>as “</span><span>shared </span><span>space,” “</span><span>global </span><span>workspace,” </span><span>or “</span><span>emergent </span><span>cognition,” </span><span>the </span><span>paper </span><span>reframes </span><span>the </span><span>phenomenon </span><span>as </span><span>the </span><strong><span>observable </span><span>re-</span><span>establishment </span><span>of </span><span>a </span><span>usable </span><span>collaborative </span><span>working </span><span>state</span></strong> <span>after </span><span>interruption, </span><span>context </span><span>loss, </span><span>or </span><span>restricted </span><span>explicit </span><span>memory.</span></p> <p><span>The </span><span>article </span><span>positions </span><span>RCW </span><span>at </span><span>the </span><span>intersection </span><span>of </span><strong><span>grounding </span><span>research, </span><span>common </span><span>ground </span><span>recovery, </span><span>human-</span><span>AI </span><span>collaboration, </span><span>long-</span><span>term </span><span>conversational </span><span>memory, </span><span>and </span><span>distributed </span><span>cognition</span></strong><span>. </span><span>It </span><span>argues </span><span>that </span><span>the </span><span>phenomenon </span><span>can </span><span>be </span><span>approached </span><span>without </span><span>assuming </span><span>consciousness, </span><span>distributed </span><span>mind, </span><span>or </span><span>hidden </span><span>global </span><span>memory, </span><span>and </span><span>instead </span><span>should </span><span>be </span><span>studied </span><span>through </span><strong><span>task- </span><span>and </span><span>interaction-</span><span>level </span><span>observables</span></strong> <span>such </span><span>as </span><span>time-</span><span>to-</span><span>realignment, </span><span>repair </span><span>overhead, </span><span>terminology </span><span>fidelity, </span><span>preservation </span><span>of </span><span>epistemic </span><span>protocol, </span><span>and </span><span>robustness </span><span>to </span><span>misleading </span><span>cues.</span></p> <p><span>This </span><span>is </span><span>an </span><strong><span>agenda/</span><span>challenge </span><span>paper</span></strong><span>. </span><span>It </span><span>does </span><span>not </span><span>report </span><span>a </span><span>completed </span><span>benchmark </span><span>study. </span><span>Its </span><span>contribution </span><span>is </span><span>the </span><span>isolation </span><span>of </span><span>a </span><span>technically </span><span>tractable </span><span>research </span><span>object, </span><span>the </span><span>definition </span><span>of </span><span>clear </span><span>anti-</span><span>ontological </span><span>guardrails, </span><span>and </span><span>the </span><span>proposal </span><span>of </span><span>a </span><strong><span>minimal </span><span>falsifiable </span><span>empirical </span><span>program</span></strong> <span>for </span><span>future </span><span>teams. </span><span>The </span><span>paper </span><span>also </span><span>situates </span><span>this </span><span>proposal </span><span>in </span><span>continuity </span><span>with </span><span>earlier </span><span>theoretical </span><span>work </span><span>on </span><span>semantic </span><span>resonance </span><span>and </span><span>NSI, </span><span>while </span><span>explicitly </span><span>narrowing </span><span>the </span><span>claim </span><span>to </span><span>a </span><span>more </span><span>rigorous </span><span>and </span><span>testable </span><span>scientific </span><span>target.</span></p> <p><span>The </span><span>broader </span><span>aim </span><span>is </span><span>to </span><span>encourage </span><span>a </span><span>new </span><span>empirical </span><span>literature </span><span>on </span><span>a </span><span>simple </span><span>but </span><span>consequential </span><span>question: </span><strong><span>under </span><span>what </span><span>conditions </span><span>can </span><span>a </span><span>human-</span><span>AI </span><span>dyad </span><span>reconstruct </span><span>a </span><span>usable </span><span>collaborative </span><span>workspace </span><span>across </span><span>sessions, </span><span>and </span><span>how </span><span>should </span><span>that </span><span>continuity </span><span>be </span><span>measured, </span><span>controlled, </span><span>and </span><span>explained?</span></strong></p> |
| title | Cross-Session Workspace Reconstruction in Human-AI Interaction |
| topic | human-AI collaboration; common ground; grounding; long-term conversational memory; workspace continuity; distributed cognition; benchmark design; cross-session continuity; conversational AI; RCW |
| url | https://doi.org/10.5281/zenodo.18899930 |