collaborators

5 papers

cs.LG2026

Forward Pass Domain Adaptation (Without Cross-Layer Backpropagation)

Rivaan Patil, Simon Dennis, Hao Guo +1

Forward-Pass-Only MLP training (FPO) adapts large language models without a backward pass through the model body, achieving 2.7--3.2x the throughput of standard fine-tuning at ~40%…

cs.AI2026

Procedural Knowledge Is Not Low-Rank: Why LoRA Fails to Internalize Multi-Step Procedures

Simon Dennis, Kevin Shabahang, Hao Guo +1

Parameter-efficient fine-tuning methods like LoRA have become the default for adapting large language models, succeeding across instruction following, style transfer, and factual a…

cs.AI2026

When Mean CE Fails: Median CE Can Better Track Language Model Quality

Hao Guo, Simon Dennis, Rivaan Patil +1

Mean cross-entropy is the standard validation metric for language models, but it can fail to track model quality during training. We examine this in two common scenarios. First, in…

cs.AI2026

Beyond Inference-Only Deployment: Comparing Weight-Based Consolidation Against Cascading Compaction

Simon Dennis, Kevin Shabahang, Hao Guo +1

Major LLM platforms deploy models in an inference-only configuration: the model serves requests but never updates per-user weights. Users must repeatedly re-teach preferences, corr…

cs.AI2026

In-Context Prompting Obsoletes Agent Orchestration for Procedural Tasks

Simon Dennis, Michael Diamond, Rivaan Patil +2

Agent orchestration frameworks -- LangGraph, CrewAI, Google ADK, OpenAI Agents SDK, and others -- place an external orchestrator above the LLM, tracking state and injecting routing…