18 papers
SAGA: Workflow-Atomic Scheduling for AI Agent Inference on GPU Clusters
Dongxin Guo, Jikun Wu, Siu Ming Yiu
AI agents execute tens to hundreds of chained LLM calls per task, yet GPU schedulers treat each call as independent, discarding gigabytes of intermediate state between steps and in…
Parameter-Efficient Neuroevolution for Diverse LLM Generation: Quality-Diversity Optimization via Prompt Embedding Evolution
Dongxin Guo, Jikun Wu, Siu Ming Yiu
Large Language Models exhibit mode collapse, producing homogeneous outputs that fail to explore valid solution spaces. We present QD-LLM, a framework for parameter-efficient neuroe…
EvoPref: Multi-Objective Evolutionary Optimization Discovers Diverse LLM Alignments Beyond Gradient Descent
Dongxin Guo, Jikun Wu, Siu Ming Yiu
Gradient-based preference optimization methods for large language model (LLM) alignment suffer from preference collapse, converging to narrow behavioral modes while neglecting pref…
The Deterministic Horizon: When Extended Reasoning Fails and Tool Delegation Becomes Necessary
Dongxin Guo, Jikun Wu, Siu Ming Yiu
Extended chain-of-thought reasoning can degrade performance on deterministic state-tracking tasks, not solely because of preference biases but, on the evidence we present, because…
Model Collapse as Cultural Evolution
Dongxin Guo, Jikun Wu, Siu Ming Yiu
Model collapse, the progressive degradation of LLMs trained on their own outputs, has been characterized statistically but lacks a linguistic explanation for which structures degra…
Do Language Models Know What Not to Say? Causal Evidence for Statistical Preemption in LLMs
Dongxin Guo, Jikun Wu, Siu Ming Yiu
How do learners acquire knowledge of what is unacceptable without negative evidence? Construction Grammar proposes statistical preemption: exposure to a conventional form (e.g., "d…
Sparse Autoencoders Map Brain-LLM Alignment onto Cortical Semantic Topography
Dongxin Guo, Jikun Wu, Siu Ming Yiu
Intermediate layers of large language models (LLMs) best predict human brain responses to language, one of the most robust findings in computational neurolinguistics, yet why remai…
Brain-LLM Alignment Tracks Training Data, Not Typology
Dongxin Guo, Jikun Wu, Siu Ming Yiu
Brain-LLM alignment is well established in English, yet the brain's language network is neuroanatomically universal across languages. Does alignment also generalize cross-linguisti…
The Deterministic Horizon: Impossibility Results as Design Specifications for Trustworthy AI Systems
Dongxin Guo
Large language models now write software, draft legal documents, and produce clinical notes, yet fundamental limits, from Turing and Arrow to the No Free Lunch theorems, shape what…
The Interference Gap: Comparing Retrieval Bounds in Human Memory and RAG Systems
Dongxin Guo, Jikun Wu, Siu-Ming Yiu
How do retrieval bounds compare between human episodic memory and Retrieval-Augmented Generation (RAG) systems under semantic interference? We present a unified signal detection th…
Bias by Necessity: Impossibility Theorems for Sequential Processing with Convergent AI and Human Validation
Jikun Wu, Dongxin Guo, Siu-Ming Yiu
Are certain cognitive biases mathematically inevitable consequences of sequential information processing? We prove that primacy effects, anchoring, and order-dependence are archite…
When Can Human-AI Teams Outperform Individuals? Tight Bounds with Impossibility Guarantees
Dongxin Guo, Jikun Wu, Siu-Ming Yiu
Human-AI teams fail to outperform their best member in 70% of studies, yet no theory specifies when complementarity is achievable. We derive tight bounds for the broad class of con…
SafeAnchor: Preventing Cumulative Safety Erosion in Continual Domain Adaptation of Large Language Models
Dongxin Guo, Jikun Wu, Siu Ming Yiu
Safety alignment in large language models is remarkably shallow: it is concentrated in the first few output tokens and reversible by fine-tuning on as few as 100 adversarial exampl…
Closing the Theory-Practice Gap in Spiking Transformers via Effective Dimension
Dongxin Guo, Jikun Wu, Siu Ming Yiu
Spiking transformers achieve competitive accuracy with conventional transformers while offering - energy efficiency on neuromorphic hardware, yet no theoretical frame…
When Do Early-Exit Networks Generalize? A PAC-Bayesian Theory of Adaptive Depth
Dongxin Guo, Jikun Wu, Siu Ming Yiu
Early-exit neural networks enable adaptive computation by allowing confident predictions to exit at intermediate layers, achieving 2-8 inference speedup. Despite widespread…
Geometric Metrics for MoE Specialization: From Fisher Information to Early Failure Detection
Dongxin Guo, Jikun Wu, Siu Ming Yiu
Expert specialization is fundamental to Mixture-of-Experts (MoE) model success, yet existing metrics (cosine similarity, routing entropy) lack theoretical grounding and yield incon…
Coalition Formation in LLM Agent Networks: Stability Analysis and Convergence Guarantees
Dongxin Guo, Jikun Wu, Siu-Ming Yiu
Large Language Model (LLM) agents are increasingly deployed in multi-agent systems requiring strategic coordination. While recent work has analyzed LLM behavior in two-player games…
Capacity-Controlled Global Attention for Graph Transformers
Yang Liu, Dongxin Guo, Tom Zheng +3
Global self-attention drives modern graph transformers, yet the softmax at its core imposes a structural constraint rarely examined directly: every attention row is non-negative an…