A comprehensive index of artificial intelligence and machine-learning research with AI-generated summaries, citation metrics, and direct links to papers and code.
Unknown
OC-PRM introduces overcredit-contrastive training to reduce false positives in process reward models, improving step-level verification precision for reasoning LLMs.
Vikash Singh, Debargha Ganguly, Aman Goel, et al.
Introduces Verdict-Preserving-Unfaithfulness (VPU) in autoformalization and proposes Generative Verification (GenV), a reference-free continuous equivalence score achieving 0.961 AUROC.
J. Jakeman, Lorena A. Barba, Joaquim R. R. A. Martins, et al.
This paper proposes a framework for verification and validation (V&V) of scientific machine learning models to enhance trustworthiness and reliability in scientific applications.
Omatharv Bharat Vaidya, Connor Thomas Jerzak, Zayne Rea Sprague, et al.
CALVER is a training-free symbolic verifier that scores LLM reasoning traces against Pearl's causal criteria, outperforming voting methods in multi-valid-answer causal reasoning tasks.
Unknown
A survey of adversarial robustness in deep learning, focusing on opportunities and challenges, including formal verification of local robustness.
Unknown
This paper proposes a hierarchical, session-isolated detection framework to identify benchmark contamination in LLM coding benchmarks, addressing solution leakage and test quality issues.
Unknown
Aurora is an automated framework for training universal process reward models using ensemble prompting and reverse verification.
Zhiqiang Liu, Wenhui Dong, Yilan Tan, et al.
TOBench is a task-oriented omni-modal benchmark for evaluating real-world tool-using agents through closed-loop multimodal verification.
Siyuan Huang, Pengyu Cheng, Haotian Liu, et al.
Skill Self-Play (Skill-SP) co-evolves a proposer, solver, and skill controller via RL to reconcile structured verification with open-ended task diversity for LLM self-evolution.
Unknown
A generative PRM trained on small synthetic data outperforms discriminative PRMs and LLM-as-a-Judge baselines in step-by-step reasoning verification.
Rahul Raguram, Ondřej Chum, Marc Pollefeys, et al.
USAC extends RANSAC with a universal framework integrating advanced sampling, verification, and model refinement for robust estimation.
Unknown
A pipeline that generates system messages and aligned assistant responses for SFT datasets lacking them, using LLM-as-a-judge verification.