OProver-32B achieves top Pass@32 scores on MiniF2F, ProverBench, and PutnamBench by combining continued pretraining with iterative agentic proving, retrieval, SFT on repairs, and RL on unresolved cases using a 6.86M-proof dataset.
Proceedings of CVPR , year=
7 Pith papers cite this work. Polarity classification is still indexing.
7
Pith papers citing it
years
2026 7representative citing papers
Cross-entropy method sampling reduces inferences needed to estimate five-nines LLM reliability by up to 156x on parameterized GSM8K templates, revealing reliability differences hidden by saturated accuracy scores.
A learnable control block trained with GRPO to select token generation order in multimodal masked diffusion models improves text-to-image alignment and multimodal understanding over logit-based heuristics.
C3RL is a new RL algorithm combining correctness, calibration, and reference accuracy rewards to improve LLM confidence calibration, enabling CAS to outperform majority voting with up to 12.33x lower inference cost.
Proposes Modality-Aware Credit Assignment (MoCA) with blindfolded-reasoning proxy to reward perception fidelity separately from reasoning in VLMs.
Introduces self-captioning and a Multimodal Interaction Gate to amplify redundant multimodal interactions, reporting 38.3% reduction in visual-induced errors and 16.8% consistency improvement.
A textbook exposition that organizes the vision-language model landscape into a modular encode–reason–decode framework and surveys architectures, losses, data, evaluation, and applications, without introducing new results.
citing papers explorer
-
OProver: A Unified Framework for Agentic Formal Theorem Proving
OProver-32B achieves top Pass@32 scores on MiniF2F, ProverBench, and PutnamBench by combining continued pretraining with iterative agentic proving, retrieval, SFT on repairs, and RL on unresolved cases using a 6.86M-proof dataset.
-
Measuring Five-Nines Reliability: Sample-Efficient LLM Evaluation in Saturated Benchmarks
Cross-entropy method sampling reduces inferences needed to estimate five-nines LLM reliability by up to 156x on parameterized GSM8K templates, revealing reliability differences hidden by saturated accuracy scores.
-
Reinforcing the Generation Order of Multimodal Masked Diffusion Models
A learnable control block trained with GRPO to select token generation order in multimodal masked diffusion models improves text-to-image alignment and multimodal understanding over logit-based heuristics.
-
Scaling with Confidence: Calibrating Confidence of LLMs for Adaptive Test Time Scaling
C3RL is a new RL algorithm combining correctness, calibration, and reference accuracy rewards to improve LLM confidence calibration, enabling CAS to outperform majority voting with up to 12.33x lower inference cost.
-
Bad Seeing or Bad Thinking? Rewarding Perception for Multimodal Reasoning
Proposes Modality-Aware Credit Assignment (MoCA) with blindfolded-reasoning proxy to reward perception fidelity separately from reasoning in VLMs.
-
Self-Captioning Multimodal Interaction Tuning: Amplifying Exploitable Redundancies for Robust Vision Language Models
Introduces self-captioning and a Multimodal Interaction Gate to amplify redundant multimodal interactions, reporting 38.3% reduction in visual-induced errors and 16.8% consistency improvement.
-
From Pixels to Prompts: Vision-Language Models
A textbook exposition that organizes the vision-language model landscape into a modular encode–reason–decode framework and surveys architectures, losses, data, evaluation, and applications, without introducing new results.