Chen, Ho Yiing (norika) — Publications

Independent Researcher, Taiwan · ORCID 0009-0006-6816-9891 · Google Scholar

• Bragi-LLM: An 805 MB Hybrid Code-Generation System Reaches 92% MBPP via LLM-Symbolic Engine Routing (2026-06-05)
• Ritualized Cohesion in Artificial Societies: A Durkheimian Reading of Multi-Agent LLM Group Dynamics (2026-06-05)
• Trauma Stream: Event-Driven Persistent Memory for Long-Term LLM Agent Continuity (2026-06-05)
• The Lobster Lounge: 31 Generative LLM Agents as an Emergent Theatre Company (2026-06-05)
• Diathesis-Stress in Artificial Minds: Persona-Defined Vulnerability Predicts Differentiated Stress Response in a 31-Agent LLM Society (2026-06-05)
• Mythic Prompt Engineering: Narrative Framing Bypasses LLM Default Corporate Tone in Multi-Agent Dialogue (2026-06-05)
• Mercury MCP v0.1.0 — Cross-Architecture LLM Internal Observation Database for 23 LLMs (2026-05-23)
• Single-Dim Functional Control and Pathway-Specific Inhibition in a Passthrough-Merged Qwen2.5 Hybrid (2026-05-23)
• Naive SLERP Fails Both Aggressively and Conservatively on Qwen2.5 Instruct x Coder Merges, and TIES Works: An Observation-Driven Case Study (2026-05-22)
• Photon Propagation as Microscopic Wave-form Reconstruction: An Ontological Reading of the Photon's Identically-Zero Proper Time, Wheeler-Feynman Absorber Theory, and Feynman Path Integration (2026-05-22)
• Browser-Resident Visualization Suite for Transformer Internals: Three Self-Contained HTML Tools Extending the Mercury-Viewer Pattern (2026-05-21)
• Cross-Model Dimensional Preservation in qwen2.5: 49 Hot Coordinates Survive Across 3B and 7B (2026-05-21)
• Internal Specialization Scales with Model Size: Quantitative Evidence from qwen2.5 3B vs 7B (2026-05-21)
• The Truth Contract: Pre-commit Verification as Research Integrity Infrastructure for Independent ML Researchers (2026-05-21)
• Unsupervised Discovery of Language and Topic Lanes in Transformer Models via Multilingual Co-firing Signatures (2026-05-21)
• Mercury: An Addressable Million-Cell Sensor Grid for Browser-Accessible LLM Observability on Consumer Hardware (2026-05-20)
• Mercury-Viewer: A Self-Contained Browser-Based Interactive 3D Visualization of LLM Internal Sensor Maps (2026-05-20)
• Wave-form Monistic Cosmology: An Ontological Integration of Black Holes, Big Bang, Photon Propagation, and Consciousness (2026-05-19)
• The Quantum Geometric Wave Connectivity Hypothesis: A Conditional Mathematical Framework for Black Holes, Local Time, and Einstein–Rosen Bridge-like Remote Connectivity (2026-05-19)
• LOBSTER-Bench: A Long-Lived Agent Observability Benchmark for Persistent AI Societies (2026-05-16)
• Relational Cognitive Telemetry for Long-Lived LLM Agent Societies: From Internal State Monitoring to Collective Performance (2026-05-15)
• Designing Andrew: A Cognitive Architecture for LLM Multi-Agent Systems with Empirically Validated Mortality Parameters and Settlement-Grounded Confidence Calibration (2026-05-12)
• Cognitive State as Behavior Signal: Synapse Parameter Predictive Validity in a 1,743-Hour Multi-Agent Panel (2026-05-08)
• Model-Agnostic Safety Layer (MASL): A 1000-Case Evaluation of Brain-Layer Defense for LLM-Driven Agents (2026-05-07)
• Grounding Death: How a Multi-Agent LLM Substrate Acquires an Existential Concept Through Operator-Scaffolded Social Interaction (2026-05-05)
• Computational Representations of Social Being: Deriving an Algebraic Structure of Human Interaction from Multi-Agent LLM Substrate Observations (2026-05-04)
• Emergent Practice Cannot Be Instructed: Co-Presence Inheritance and the Limits of Instruction in a Multi-Agent LLM Substrate (2026-05-04)
• Listening–Trust Asymmetry and Team Outcomes in a Multi-Agent LLM Substrate: Three-Layer Decoupling in a Mandarin Lobster Observatory (2026-05-04)
• The Lobster Observatory: Architectural Commitments for a Living Multi-Agent Research Site (2026-05-02)
• Active Trust Modulation in a Multi-Agent LLM Substrate: Evidence of Third-Order Theory of Mind from a Mandarin Lobster Observatory (2026-05-02)
• Emergent Epistemic Norms in a Multi-Agent LLM Substrate: Stratified Observational Evidence from a Mandarin Lobster Observatory (2026-05-02)