REVIEW 6 cited by
PairJudge RM: Perform Best-of-N Sampling with Knockout Tournament
Not yet reviewed by Pith; the record is open.
This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.
SPECIMEN: schema-true, not a live event
T0 review · schema-true
One-sentence machine reading of the paper's core claim.
pith:XXXXXXXX · record.json · timestamp
PairJudge RM: Perform Best-of-N Sampling with Knockout Tournament
read the original abstract
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 Judge Reward Model (PariJudge RM) combined with a knockout tournament for BoN sampling. Instead of assigning absolute scores, given one math problem, PariJudge RM judges two candidate solutions' correctness with chain-of-thought reasoning simultaneously. This approach eliminates the need for scoring and enables cross-validation of solutions through parallel judgment. In the knockout tournament, PariJudge RM conducts pairwise Judgment between candidate solutions and eliminates the incorrect ones iteratively. We construct PairJudge-432K, a large-scale dataset of 432K pairwise judgments derived from NumiaMath and annotated using \texttt{gemini-1.5-flash}, and train the PariJudge RM via supervised fine-tuning. Experiments on MATH-500 and the Olympiad Bench demonstrate significant improvements over baseline reward models. And a 40\% to 60\% relative improvement is achieved on the top 50\% challenging problems.
Forward citations
Cited by 6 Pith papers
-
Joint Consistency: A Unified Test-Time Aggregation Framework via Energy Minimization
Joint Consistency casts test-time aggregation as Ising-type energy minimization with pairwise LLM-judge interactions, subsuming voting methods and outperforming baselines across reasoning tasks.
-
Z-Reward: Beyond Scalar Rewards by Internalizing Reasoning into Score Distributions
Z-Reward trains a 27B reasoning teacher VLM on score distributions via GDSO and distills it via RISD into a 9B student, reaching 89.6% and 88.6% human preference accuracy with 41.3% optimization gain over SFT baseline.
-
Z-Reward: Beyond Scalar Rewards by Internalizing Reasoning into Score Distributions
A teacher-student reward model learns reasoning-conditioned score distributions for text-to-image images, yielding ~89% preference accuracy and a 41% net human-preference gain when used for generator optimization.
-
Scalable Token-Level Hallucination Detection in Large Language Models
TokenHD uses a scalable data synthesis engine and importance-weighted training to create token-level hallucination detectors that work on free-form text and scale from 0.6B to 8B parameters, outperforming larger reaso...
-
Scaling Test-Time Compute to Achieve IOI Gold Medal with Open-Weight Models
GenCluster scales test-time compute via large-scale generation, behavioral clustering, ranking, and round-robin submission to achieve IOI gold medal performance with the open-weight gpt-oss-120b model.
-
Reinforcement Learning from Human Feedback: A Statistical Perspective
A statistical survey of RLHF for LLM alignment that connects preference learning and policy optimization to models like Bradley-Terry-Luce while reviewing methods, extensions, and open challenges.
discussion (0)
Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.