Cross-Scale Alignment Consistency of the BNRM Framework Versus Bradley-Terry Models on the Chatbot Arena Leaderboard

Abstract

Abstract: Foundation models (FMs), particularly large language models (LLMs), have shown significant promise in various software engineering (SE) tasks, including code generation, debugging, and requirement refinement. Despite these advances, existing evaluation frameworks are insufficient for assessing model performance in iterative, context-rich workflows characteristic of SE activities. To address this limitation, we introduce emph\SWE-Arena\, an interactive platform designed to evaluate FMs in SE tasks. SWE-Arena provides a transparent, open-source leaderboard, supports multi-round conversational w

Research Question

Does the BNRM framework maintain consistent alignment performance across different language model scales (e.g., Qwen2.5-7B vs. Qwen2.5-72B) when evaluated using the Chatbot Arena leaderboard, and how does this compare to standard Bradley-Terry models?

Verification Level

Descriptive public verification status only; aggregate claim counts are public, but individual claim records are not exposed here.

Truth-Engine Gate Verdict

This record has not completed Gate 2 of the verification pipeline (a type-checked Lean4 proof for mathematical claims, or a sealed-sandbox reproduction for empirical claims). It is a literature synthesis only. VERIFIED requires an attached reproducible artifact (Lean4 proof source, or repro script and results) before this status can be set; it is not derived from review score or claim count.

Quality Tier

Descriptive public triage only; this tier does not alter current publication or DOI behavior.

Quality Dimensions

Automated triage signals derived from public fields; not human peer review or independent validation.

Correction Record

Public corrections are additive records. Current status does not claim the synthesis is error-free.

Provenance