IntegriRef
Accepted at NLPCC 2026. The first citation verification system that goes beyond existence checking to full semantic cla…
Building the future, one commit at a time. Here are some of my open-source projects and research work.
Accepted at NLPCC 2026. The first citation verification system that goes beyond existence checking to full semantic cla…
ML-aware caching for recommendation systems: 45% cache hit rate, 85% latency reduction on hits, and under 3% NDCG degra…
Quotient space reasoning for LLMs: identifies 64-72% semantic redundancy and reduces token usage by 26.8%.
KV cache optimization achieving 10-16x faster warm latency vs vLLM, validated across 84 tests on TinyLlama, Mistral-7B,…
A systematic security study formalizing Persistent Semantic Entities (PSEs) as a new attack surface in tool-augmented L…
Mechanism-designed belief aggregation for edge AI: VCG-based truthful elicitation achieves lower false negative rates t…
An energy-based / density-based risk modeling framework that challenges the anomaly=risk assumption in ML: GMM density …
The largest empirical study of MARL coordination strategies applied to LLM multi-agent systems: 3,400+ episodes across …
A 465M-parameter fine-tuned router that classifies queries into 7 distinct epistemic universes (frequentist, Bayesian, …