arXiv: "Does Reasoning Make Language Models Agree?" by Gorecki & Hardt
rohanpaul_ai · x · 2026-11-05
arXiv page (2610.41691, submitted 2026-10-30) for the same Max Planck paper.
- Authors Mila Gorecki and Moritz Hardt, Max Planck Institute for Intelligent Systems and Tübingen AI Center
- Key finding: enabling reasoning increases inter-model agreement, while individuals eliciting longer reasoning traces get less consistent predictions; both patterns persist after controlling for accuracy
- Asking models to reason more promotes agreement; observing longer reasoning signals disagreement
- Listed under cs.LG / cs.CL / cs.CY
Same paper as the companion post in this batch; see that entry for detailed numbers.
Related event: Max Planck Paper: Reasoning Makes LLMs Converge, Even on Errors(2 posts)→
More from Research
- Open-source segmenter tracks a breathing tumor across 10 CT phases on one GPU — MaziyarPanahi · 2026-11-05
- Mathematician: AI can grind lemmas, but 100-page arXiv proofs must ship with full Lean formalization — LucaAmb · 2026-11-05
- Bregman vantage point trees: VP-trees using Bregman balls for fast nearest-neighbor queries — FrnkNlsn · 2026-11-05
- Pre-print: iterated creation, grading, and revision of data science projects by language models — cjmaddison · 2026-11-05
- Latent Sparse Attention on Apple Silicon: 31% faster TTFT at 128k with a hybrid prefill/decode split — GabGarrett · 2026-11-05
- Cursor & UC Berkeley's Targeted Guidance teaches RL new behaviors without degrading baselines — burny_tech · 2026-11-05