arXiv cs.AIPaper
Not All Test-Time Compute Is Evidence: Model-Specific Bias in LLM Reasoning
This cracks a real problem in scaling test-time compute: just running the same model longer doesn't give you more signal. DIA is elegant: disagreement between intermediate and final outputs flags reasoning instability, then you get a second opinion from different models, not more steps from the same one. It's 7% gain on MuSiQue while using 47% fewer tokens than brute-force multi-agent, which is the kind of efficiency that matters. If you're building inference-time scaling, this is your read.