Aggregating many LLM-synthesized weak verifiers via weak learning from sparse labels yields stronger verifiers that improve F1 by up to 7X over direct LLM judges on 3D room and 2D poster tasks and boost generation quality by 66.2%.
The answer is (X)\
6 Pith papers cite this work. Polarity classification is still indexing.
6
Pith papers citing it
citation-role summary
background 1
citation-polarity summary
roles
background 1polarities
background 1representative citing papers
Small 7B reasoning models were fine-tuned on synthetic and curated QFT problems using RL and SFT, yielding performance gains, error analysis, and public release of data and traces.
FUSE ensembles verifiers unsupervisedly by controlling their conditional dependencies to improve spectral ensembling algorithms, matching or exceeding semi-supervised baselines on benchmarks including GPQA Diamond and Humanity's Last Exam.
Expecting over scoring-token logits yields continuous, scalable verification that improves agent trajectory selection and dense RL rewards across coding, robotics, and medical benchmarks.
Closed-loop self-evolution on LLMs improves reasoning on Knights and Knaves tasks but plateaus short of oracle-supervised levels, with multi-turn revision nearly matching it for large models.
Empirical tests across 12 benchmarks and 14 models show verifiers certify easy problems more reliably, catch errors from weak generators more easily, and correlate with their own solving skill in a difficulty-dependent way.
citing papers explorer
-
Aggregating LLM-Based Weak Verifiers for Spatial Layout Generation
Aggregating many LLM-synthesized weak verifiers via weak learning from sparse labels yields stronger verifiers that improve F1 by up to 7X over direct LLM judges on 3D room and 2D poster tasks and boost generation quality by 66.2%.
-
Fine-Tuning Small Reasoning Models for Quantum Field Theory
Small 7B reasoning models were fine-tuned on synthetic and curated QFT problems using RL and SFT, yielding performance gains, error analysis, and public release of data and traces.
-
FUSE: Ensembling Verifiers with Zero Labeled Data
FUSE ensembles verifiers unsupervisedly by controlling their conditional dependencies to improve spectral ensembling algorithms, matching or exceeding semi-supervised baselines on benchmarks including GPQA Diamond and Humanity's Last Exam.
-
LLM-as-a-Verifier: A General-Purpose Verification Framework
Expecting over scoring-token logits yields continuous, scalable verification that improves agent trajectory selection and dense RL rewards across coding, robotics, and medical benchmarks.
-
On the Generalization Gap in Self-Evolving Language Model Reasoning
Closed-loop self-evolution on LLMs improves reasoning on Knights and Knaves tasks but plateaus short of oracle-supervised levels, with multi-turn revision nearly matching it for large models.
-
Variation in Verification: Understanding Verification Dynamics in Large Language Models
Empirical tests across 12 benchmarks and 14 models show verifiers certify easy problems more reliably, catch errors from weak generators more easily, and correlate with their own solving skill in a difficulty-dependent way.