{"slug": "from-discovery-to-decision-finite-budget-recoverability-in-llm-voting", "title": "From Discovery to Decision: Finite-Budget Recoverability in LLM Voting", "summary": "A new arXiv paper (2610.01014v1) derives a sharp recoverability threshold for LLM voting under a fixed call budget, characterizing a \"discovery-to-decision gap\" in which a correct answer found early can still fail to accumulate enough support to win the final plurality. In a controlled Word16 study, input permutation improved raw-plurality accuracy by 21.1 points with essentially unchanged single-call correctness, and the authors' gold-free exact locking certificate saved 28-30% of calls at a 16-call budget while preserving every fixed-budget output. The work also shows that merging wrong-answer identities preserves single-call correctness and cannot improve plurality accuracy.", "body_md": "arXiv:2610.01014v1 Announce Type: new \nAbstract: Voting over multiple LLM responses is a common primitive in test-time scaling and ensemble inference. Collecting more responses can expand the candidate pool and increase the chance that a correct answer is discovered. Under a fixed call budget, a discovered answer still needs to accumulate enough support within the remaining calls to become the final plurality winner, creating a discovery-to-decision gap. In this work, we characterize this gap through the realized vote state and remaining call budget. We derive a sharp recoverability threshold and show that, as sampling proceeds, the observed candidate set can only expand while the set of reachable endpoint winners can only contract, inducing a candidate-level conversion window. Under a specified iid response law, the same state yields exact finite-horizon endpoint probabilities. We further show that merging wrong-answer identities preserves single-call correctness and cannot improve plurality accuracy, and that the effect of redistributing wrong-answer probability depends on the realized vote state. Singleton reachability yields a gold-free exact locking certificate. For a known answer universe, its first trigger is the earliest prefix at which all admissible continuations yield the same fixed-budget output. Empirically, most discovered-but-unselected correct answers lose reachability only after discovery. In a controlled Word16 study, input permutation improves raw-plurality accuracy by 21.1 points with essentially unchanged single-call correctness. Exact locking saves 28-30% of calls at a 16-call budget while preserving every fixed-budget output.", "url": "https://wpnews.pro/news/from-discovery-to-decision-finite-budget-recoverability-in-llm-voting", "canonical_source": "https://www.machinebrief.com/news/from-discovery-to-decision-finite-budget-recoverability-in-l-rcb8", "published_at": "2026-10-02 04:00:00+00:00", "updated_at": "2026-10-02 04:45:45.329982+00:00", "lang": "en", "topics": ["large-language-models", "ai-research", "machine-learning", "artificial-intelligence"], "entities": ["arXiv", "Word16"], "also_reported_by": [], "alternates": {"html": "https://wpnews.pro/news/from-discovery-to-decision-finite-budget-recoverability-in-llm-voting", "markdown": "https://wpnews.pro/news/from-discovery-to-decision-finite-budget-recoverability-in-llm-voting.md", "text": "https://wpnews.pro/news/from-discovery-to-decision-finite-budget-recoverability-in-llm-voting.txt", "jsonld": "https://wpnews.pro/news/from-discovery-to-decision-finite-budget-recoverability-in-llm-voting.jsonld"}}