knockout tournament
GenSelect: A Generative Approach to Best-of-N
Toshniwal, Shubham, Sorokin, Ivan, Ficek, Aleksander, Moshkov, Ivan, Gitman, Igor
Generative reward models with parallel sampling have enabled effective test-time scaling for reasoning tasks. Current approaches employ pointwise scoring of individual solutions or pairwise comparisons. However, pointwise methods underutilize LLMs' comparative abilities, while pairwise methods scale inefficiently with larger sampling budgets. We introduce GenSelect, where the LLM uses long reasoning to select the best solution among N candidates. This leverages LLMs' comparative strengths while scaling efficiently across parallel sampling budgets. For math reasoning, we demonstrate that reasoning models, such as QwQ and DeepSeek-R1-0528, excel at GenSelect, outperforming existing scoring approaches with simple prompting.
Pairwise RM: Perform Best-of-N Sampling with Knockout Tournament
Liu, Yantao, Yao, Zijun, Min, Rui, Cao, Yixin, Hou, Lei, Li, Juanzi
Best-of-N (BoN) sampling, a common strategy for test-time scaling of Large Language Models (LLMs), relies on reward models to select the best candidate solution from multiple generations. However, traditional reward models often assign arbitrary and inconsistent scores, limiting their effectiveness. To address this, we propose a Pairwise Reward Model (Pairwise RM) combined with a knockout tournament for BoN sampling. Instead of assigning absolute scores, given one math problem, Pairwise RM evaluates two candidate solutions' correctness simultaneously. This approach eliminates the need for arbitrary scoring and enables cross-validation of solutions through parallel comparison. In the knockout tournament, Pairwise RM conducts pairwise comparisons between candidate solutions and eliminates the incorrect ones iteratively. We construct \ourdataset, a large-scale dataset of 443K pairwise comparisons derived from NumiaMath and annotated using \texttt{gemini-1.5-flash}, and train the Pairwise RM via supervised fine-tuning. Experiments on MATH-500 and the Olympiad Bench demonstrate significant improvements over traditional discriminative reward models. And a 40\% to 60\% relative improvement is achieved on the top 50\% challenging problems.
Merging Knockout and Round-Robin Tournaments: A Flexible Linear Elimination Tournament Design
We propose a new tournament structure that combines the popular knockout tournaments and the round-robin tournaments. As opposed to the extremes of divisive elimination and no elimination, our tournament aims to eliminate the participants as linearly as possible as a form of subtractive elimination. Our design is flexible in the sense that it can be adapted to any number of players $N$ and any number of matches $M$. Our design satisfies many properties that are desirable for a tournament to select a winner and can be adapted to rank all the participating players.
Rigging Nearly Acyclic Tournaments Is Fixed-Parameter Tractable
Ramanujan, M. S. (Technische Universität Wien) | Szeider, Stefan (Technische Universität Wien)
Single-elimination tournaments (or knockout tournaments) are a popular format in sports competitions that is also widely used for decision making and elections. In this paper we study the algorithmic problem of manipulating the outcome of a tournament. More specifically, we study the problem of finding a seeding of the players such that a certain player wins the resulting tournament. The problem is known to be NP-hard in general. In this paper we present an algorithm for this problem that exploits structural restrictions on the tournament. More specifically, we establish that the problem is fixed-parameter tractable when parameterized by the size of a smallest feedback arc set of the tournament (interpreting the tournament as an oriented complete graph). This is a natural parameter because most problems on tournaments (including this one) are either trivial or easily solvable on acyclic tournaments, leading to the question — what about nearly acyclic tournaments or tournaments with a small feedback arc set? Our result significantly improves upon a recent algorithm by Aziz et al. (2014) whose running time is bounded by an exponential function where the size of a smallest feedback arc set appears in the exponent and the base is the number of players.
Fixing a Balanced Knockout Tournament
Aziz, Haris (NICTA and UNSW) | Gaspers, Serge (NICTA and UNSW) | Mackenzie, Simon (NICTA and UNSW) | Mattei, Nicholas (NICTA and UNSW) | Stursberg, Paul (TU Munich) | Walsh, Toby (NICTA and UNSW)
Balanced knockout tournaments are one of the most common formats for sports competitions, and are also used in elections and decision-making. We consider the computational problem of finding the optimal draw for a particular player in such a tournament. The problem has generated considerable research within AI in recent years. We prove that checking whether there exists a draw in which a player wins is NP-complete, thereby settling an outstanding open problem. Our main result has a number of interesting implications on related counting and approximation problems. We present a memoization-based algorithm for the problem that is faster than previous approaches. Moreover, we highlight two natural cases that can be solved in polynomial time. All of our results also hold for the more general problem of counting the number of draws in which a given player is the winner.